TWI302358B - Method and system for determining optical properties of semiconductor wafers - Google Patents

Description

1302358 IX. Description of the invention: [Technical field to which the invention pertains] Month 5 = control system and method. This case advocates the provisional case of the Wu Guozhong gyue case number 6_6, _ of the 5th and 5th of July. [Prior Art]

Both industry and scientific processes report on the correct measurement of the surface temperature of hot objects. In the case of the element, the temperature must be measured and controlled. More specifically, the temperature of the semiconductor wafer must be accurately monitored during rapid thermal oxidation or other improvements or additions to the wafer surface. In the above system (4), it should be determined that the substrate temperature is accurate between _ dryness, and the range may extend from less than 400 degrees Celsius to more than Celsius. " The temperature measurement of high-temperature objects is not based on (1) contact method, or (2) non-contact method. For example, if the contact method is used, the high temperature object is in contact with the sensor, such as a "thermocouple," which in turn is connected to a thermometer that indicates the temperature of the object. On the other hand, the conventional non-contact s-temperature method includes using a light sensor, such as an optical pyrometer, to sense a certain wavelength of light emitted by the object. If the thermal radiation from the object is known, the temperature of the object can be estimated. To deal with semiconductor materials in the electronic industry, it is generally preferred to measure the semiconductor wafer temperature by the contact method. For example, the advantage of the non-contact method is that the heat treatment _ wafer may rotate. The ram is to promote uniform temperature distribution across the wafer. Laying the wafer also promotes processing gas and contact with the wafer. In addition to being able to rotate the wafer, the advantage of the non-contact method is that it does not require any thermometer attached to the wafer, and the wafer can add valuable time to the semiconductor fabrication process more quickly. One of the most important requirements for the foreseeable wafer process is the ability to measure crystal_true temperature with high accuracy, reproducibility and speed. Accurate L's ability to measure wafer temperature, directly in the quality and size of the fabricated semiconductor components

C:\Eunice 2006\PU CASE\PU-〇68\PU^ 068-00 ΐ3\ρυ.( 〇68-〇〇13. TSUEI.D0C 1302358 Calculation speed. Type of energy (4) minimum crane size limit The microchip is off. Therefore, there is a control system for the ability to measure and control temperature during the process of the component. The industry is gradually pressured to develop more accurate temperature measurements and the disadvantage is that the traditional non-contact optical pyrometer The system's main surface is more realistic than the apparent temperature. In particular, a solid object (such as, for example, a shot) can calculate the temperature of the black body. However, an actual shot - part. The emitted radiation is only the radiation emitted by the light object emitted by the black body at the same temperature: t 2 The emissivity of the actual object. Therefore, the sensing is caused by the actual temperature. ^Tenton* shows a view that is not the true temperature of the object. Because it is 匕, 取—| should be repaired according to its emissivity. 'The true temperature of the object in the ten-measurement, the temperature displayed is difficult to accurately measure. Unfortunately, the actual _ emissivity is usually unknown. , also the characteristics of the circle (the half-day yen's rate is different) The Wei rate is the coarseness of the surface of the crystal. The chemical composition of the wafer, the thickness of the wafer, and any coating on the wafer a Ba® have the wavelength used by the pyrometer. Wavelength:: ί:月One::: Emissivity of bulk wafers' If the wafer is used by a pyrometer, {there is also difficulty in properly measuring the wafer temperature. For example, the method should measure its characteristics before or during wafer processing. The reflectance is disclosed in a method of facilitating, in which the semiconductor wafer can be measured for the spear circle of the invention: emissivity. The semiconducting process processed by the high temperature process chamber, such as the optical property of the substrate, such as crystal The many characteristics of the circle are the Days of the Day®. The method is measured according to the method disclosed by the present invention. The tree is used to better control the heating process and/or the substrate is heated.

6 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.D〇C 1302358 SUMMARY OF THE INVENTION The present invention is generally related to a kind of measurement_substrate (like The method of optical characterization of the semiconductor wafers' to heat the wafer more accurately during the thermal process, or to better control a plurality of different pure component neutrons during the heating process. The methods and systems disclosed herein can improve wafer temperature readings and their accuracy by-radiation sensors (e.g., pyrometers) or can improve the thermal radiation characteristics of substrates and/or estimates. In a specific implementation, the substrate optical body produced by the method can be provided to the controller to control the temperature of the circle.

For example, in a particular embodiment, the present invention is directed to a method of measuring at least one filament characteristic of a semiconductor crystal. This method includes the first surface of the semiconductor wafer. The aperture emitted onto the first surface of the semiconductor wafer is introduced into an optical path configured to reflect the light reflected by the first surface and penetrate the wafer from the other side of the wafer. The light reflected by the second surface is separated. The light reflected by the first surface is separated from the light reflected by the second surface, and the light reflected by the first surface can be detected by a detector. For example, the detector can be any photosensor and can be configured to reflect the wavelength of the light on the surface of the pomelo. According to the present invention, at least the optical characteristics of the semiconductor crystal can be measured based on the light reflected by the first surface. This property may include the reflectance of the first surface, the refractive index of the surface, the absorption of the first surface, or the transmittance of the first surface. In addition, the optical properties may include reflectivity, emissivity, and absorptivity of the semiconductor wafer. Furthermore, in addition to determining at least one optical property of the first surface of the wafer, the method can also be used to determine at least the optical properties of the day-to-day. The light reflected by the first surface is separated from the inverted light of the other surface of the wafer by the wearer, and the period of the female can be different. For example, the optical path can include a recording optical component. These optical elements can include mirrors, lenses, apertures, and the like. For example, in a particular embodiment, the first lens and the second lens direct light to the first surface of the semiconductor wafer C:\Eunice 2006\PU CASE\PU-〇ea\Pu 068 -0013\PU-068-0013- TSUEI. Doc 1302358. The light is reflected by the first surface, which in turn passes through the second lens. With this second lens, the light is reflected by a mirror and passes through the third lens for focusing to a light sensor. However, it is conceivable that the above specific embodiments show only one example of the optical path that can be used in the present invention. The manner in which the light reflected by the first surface is distinguished from the reflected light of the second surface when the light travels through the optical path may vary from application to application. For example, different beams can be distinguished by adjusting the focal length of one or more lenses within the system. Additionally or additionally, the system can include a plurality of different apertures or filters to distinguish between different beams. In another embodiment, the angle of incidence of the emitted light onto the first surface of the semiconductor wafer can be distributed to separate the first surface reflected light from the second surface reflected light.

The source of light used to emit light onto the first surface of the substrate may depend on the particular application. For example, in one embodiment, the light can comprise a broadband source. Alternatively, the source emits a laser beam. Once at least one optical characteristic of the semiconductor wafer is determined as described above, the optical characteristics are utilized and incorporated into different systems and processes. For example, in one embodiment, one or more optical characteristics are utilized to control the heating process for the semiconductor wafer. In this embodiment, depending on the optical characteristics, at least one of the system components can be controlled during the process of heating the semiconductor. For example, in one embodiment, the system component can include a temperature measurement system that includes a dynamometer measuring component 'a pyrometer—a sensor that senses the semiconductor wafer to emit _radiation during heating. The semiconductor wafer temperature is determined. The amount of light from the first side to the surface can determine the emissivity of the semiconductor wafer. This emissivity is matched with the amount of (four) sensed by the spying device, which can be used to deceive the semiconductor wafer temperature. In the embodiment of the method, the sensitization senses the amount of light reflected by the first surface of the semiconductor wafer emitted by the semiconductor wafer, and the sensation of the sensation is visibly. Made with the same wavelength. The measurement of the amount of light reflected by the semiconductor substrate also emits light 'from the first surface of the semiconductor wafer to the operating wavelength of the illuminating sensing element. The a=== rate is more ^ 8 C:\Eunice 2006\PU CASE\ PU-068\PU-068-W13\PU-068-0013-TSUEI.Doc 1302358 degrees:, the power controller can be adjusted. For example, this pure = derivative f: a mixture of both heated & denier, / 疋. The _ to the surface of the semiconductor wafer can be used to determine that the semiconductor wafer is in the boost period controller and is thus optionally increased or decreased; it is set by power or energy. In this embodiment, the surface of the light is reflected, and the range of wavelengths of the electromagnetic radiation used to detect the wavelength range of the semiconductor and/or the heated wafer overlaps. At the time of soil temperature, the specific semiconductor wafer optics can be at a low level. In this case, the amount of light reflected by the surface of the first surface of the wafer is 八 ι 表 表 表 表 表 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射 反射The diameter is such that the light reflected from the first surface of the wafer is separated from the crystal, and the static (10) light. The obtained anti-scale is re-exposed to the transmittance of the semiconductor wafer when the transmittance is greater than the temperature of α1 at the wavelength of the radiation-induced if?1. The transmittance and emissivity thus determined can then be used to calibrate the temperature measurements obtained with the radiation sensing device. 4 = f The method disclosed in the present invention can also be used to control the electrical level of the heating device at low temperatures. However, in this embodiment, the light reflected from the first surface of the semiconductor wafer by the A, 'line and the second surface of the semiconductor wafer is used to detect the wavelength j and the substrate used to heat the wafer. The wavelength range of electromagnetic radiation is substantially overlapping. In this way, the absorption rate can be determined and used to optimize the power or energy setting. = The optical properties of the conductor wafer can be determined in the thermal process chamber as described above, or in the process chamber. For example, in one embodiment, the optical properties can be measured at any suitable location. For example, it can be measured at a location on the robotic arm or in a separate process chamber. Certainly measured 9 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013 TSUEI.Doc 1302358 The above optical characteristic process is obtained: the viewer can see the crystal Other changes and perspectives of the invention will be discussed in detail below. [Embodiment] 2. Those skilled in the art should be able to understand the description of this specification, and are not intended to limit the broader viewpoint of the present invention. In general, the present invention relates to a system for controlling the optical characteristics of optical characteristics that can be used to determine the beauty and then the operation. For example, the system of the private order of the class (4) In one embodiment, the substrate can include a semiconductor am, which is better to measure and control the wafer. The first sub-characteristic can be used to control heating of the wafer. The method disclosed in the present invention can be applied to any substrate other than a semiconductor wafer, such as a forged tape, a rubber fiber, a thread, and the like. The slab, including the semiconductor material, may be used in the heat treatment of the substrate, such as the oxidation of the substrate, or other processes in the substrate surface modified or added _. Other processes used in the invention include any suitable film, such as a vapor phase vapor deposition process or an atomic layer deposition process. The principles of the present invention are used during a plasma process in which a material or a substrate is to be deposited on a substrate. . According to an exemplary embodiment of the first figure, a system (10) is shown for use in the process of the present invention: a substrate (such as a semiconductor-based wafer). The system (1) includes a process chamber (12), The second modification can accommodate substrates such as a wafer (14) for performing various processes. The wafer (on the substrate carrier (丨5) of the process chamber (12), the latter can optionally pass The configuration is capable of rotating the dome. The process chamber (12) is designed to heat the wafer (14) at a very fast rate and under carefully controlled conditions. The process chamber (12) can be made from a variety of materials, including Some metals, glass, and ceramics. For example, the process chamber (12) can be made of stainless steel or quartz. 10 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSU£ I.Doc 1302358 If the process chamber (12) is made of a thermally conductive material, the process chamber may include a cooling system. For example, in the first figure, the process chamber (10) includes a cooling tube (10) surrounding the "process The outer edge of the crucible. The cooling tube (16) is suitable for circulating a coolant (like water), which can be used to maintain the temperature of the outer wall of the system. An intake port (18) and an exhaust port (2〇) may be included to introduce gas into the chamber and/or maintain the pressure in the chamber within a predetermined range. For example, it may be via a two-port (1S) Introducing a gas into the process chamber (12) and the wafer (Μ) to react. Once processed, the gas can then be vented to the process chamber using an exhaust port (18).疋

Alternatively, an blunt gas can be fed into the process chamber (12) via the air inlet (18) to avoid unwanted or undesirable side reactions within the chamber. In still further embodiments, the intake port (^) and the exhaust port (2〇) can be used to pressurize the process chamber (10). A vacuum can be created in the process chamber (10) if desired. ° During processing, the process chamber (12) can be modified to rotate the wafer (14) using a wafer rotation mechanism. Rotating the wafer promotes a more uniform temperature across the wafer surface and can also help to enhance gas contact between the wafer (14) and any introduced process chamber. It is conceivable that in addition to the semiconductor wafer process chamber, it is also possible to retrofit optical components, fine, fiber, forged and other substrates of any particular shape. The air can be equipped with one or more heating devices to heat the wafer ill during the process. In this embodiment, a heating device (22) includes a plurality of luminaires (24), such as crepe halogen = and fox light 4, a laser or a mixture of the above. The heating means (9) may comprise a reflective iridium or group of mirrors for directing the thermal energy emitted by the heating means to the wafer (4). As shown in Fig. 1, the luminaire (24) can be placed above the wafer (10). However, it is conceivable that the luminaire can be placed in a granule. For example, the system (1_ may include additional luminaires, and is placed not only above the wafer (14) but also under the wafer (14). As an alternative to using a luminary or in addition to multiple luminaires a variety of other heating devices, for example, the heating device can be configured to add two deniers. Any suitable electromagnetic light. For example, the heating device also emits radio frequency or micro/month b. In his embodiment, The substrate can be heated in a hot wall environment or via conduction. The energy beam can also be used in the base 11 C:\Eumce 2006\PU CASE\P (J-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 For example, the energy beam can include a beam, an electron beam, or an ion beam. In a particular embodiment, the process chamber can include a heater block. For example, a heater block can be placed over or under the wafer to For heating crystals _ not in contact with the wafer. Such heating seats have been well-known in the industry. As shown in the first figure, one or more heating devices, such as lamps (24), can be equipped with - progressive power control The device (25) 'can be used to increase or decrease the thermal energy emitted by the heating device.

The thermal process chamber (10) further includes a plurality of optical fibers or light pipes (28) which in turn are in communication with a plurality of corresponding photodetectors (3 turns). The fiber (28) is configured to receive thermal energy emitted by the wafer (14) at a particular wavelength. The detected amount of radiation is then coupled to an optical device (10) which produces a - (iv) voltage signal for the temperature of the wafer to be used to smear the wafer (28) and the optical device (10) to form a pyrometer. Bei Guanzhong mother performs the process program, and the system (10) can be designed so that the fiber is only fired by the crystal_), not by the radiation emitted by the lamp (24). In this regard, the system (10) = bracket - filter, mirror (32) to avoid the thermal detector of the wavelength of the light detector (3 〇) emitted by the luminaire into the == (12). The filter (32) may be the -f port, which in the embodiment is composed of the molten stone stone. As described above, the semiconductor wafer (14) is measured by the fiber (28) and the side () during the process. More specifically, the photodetector (10) senses a particular $long shot emitted by the wafer (14) to determine the temperature. In order to accurately calculate the amount of radiation according to the optical scale (10), it is necessary to know or otherwise estimate a plurality of characteristics of the wafer (14). For example, the temperature 疋疋 is based on the reflectance, transmittance, and (e) emissivity of the crystal K(10), which are difficult to predict or estimate. These values will vary not only by the temperature of the wafer, but also by any configuration that may be made on the wafer (14) during the process. There have been many attempts to design a non-contact temperature measurement system that can be used for the characteristics of the wafer (14). For example, in some embodiments, in wafer processing / a ', | or as measured. For example, U.S. Patent No. 6,056,434 (incorporated herein by reference) C:\Eunice 2〇〇6\PU CASE\PU 12 _m_cm〇〇mpu side. Coffee tsuei d〇c 1302358 Discussion of the local temperature measurement using a reflectometer . In other embodiments, there are also variations in the various properties of the substrate to be measured prior to the process. However, as mentioned above, the optical properties of the substrate can vary with temperature. Moreover, some of the materials on the substrate may be transformed or other materials formed on the substrate, altering the construction and optical properties of the substrate. These barriers have limited the ability to use pre-process characterization as a way to improve the process.

One of the main questions asked is that the optical absorption of semiconductor materials (such as Shi Xi) is affected by its temperature and the exchange of impurities. For example, a lightly doped wafer is generally semi-permeable at room temperature to wavelengths greater than about 1.1 μηη. Therefore, the general measurement carried out by the company at room temperature is affected by the light that penetrates the substrate and is reflected by the crystal. When the wafer is heated in the process system, the _ absorption coefficient rises rapidly with temperature and the crystal becomes more transparent. This change causes the reflectivity of the wafer to change. In this case, the room temperature measurement of the characteristics can not be used to improve the temperature; the present invention relates to the measurement of the substrate before the addition of the wafer - the substrate (such as the body, the optical miscellaneous method and system) According to the present invention, "measuring by other methods, the various optical characteristics of the wafer can be used not only to assist in obtaining a more accurate measurement of temperature production" but also to optimize the heat absorption. The information obtained by clearing the branches allows the temperature measurement not only to be carried out at low temperatures, but also to perform the method. The method of the gamma can be carried out outside the curing process. Alternatively, the temperature 敎 can also be applied at the time of the health. Before: In this case, a brief description of how the beams interact will be very good—Figure 4 is not representative of a substrate such as a wafer, which is illuminated by an incident angle θ of two = Α (). Top surface (WF) and bottom surface (WB) ° Ϊ Ϊ Ϊ elemental touch, which can affect the wire _ reflectivity and transmission ^ ^ ^ ^ some _ is _ surface reflection, forming a reflection of light illusion. The second part of the tooth ^ The surface (WF)' forms internal light A1. This light is at a different angle, 'this is because of the day The difference between the refractive index difference between the Japanese yen and the human-induced medium containing the ray A0 is 13 C: (10) 吆 should be the result of the UEI.Doc 1302358. When the light A1 travels through The thickness of the wafer may be weakened by the absorption of energy within the wafer. Typically, this absorption is exponentially dependent on the length of the path through the wafer, the so-called "Beer, sLaw". A1 reaches the back of the wafer, and a portion of the light is transmitted through the surface to form a light ray. The second portion is reflected by the back surface (WB) to form a second internal light, A2. If the back side of the wafer WB arrives at A1, it is the original The front surface of the wafer to which the light A is incident is parallel. As long as the medium outside the back of the wafer has the same refractive index as the medium containing A0, the light T1 will travel in the direction parallel to the original light A0. The light A2 will also A1 is the same angle as the vertical line of the wafer. A2 is also absorbed and weakened as it travels toward the front surface wf of the wafer. Some of it will be transmitted to form light R2, and part of it will be reflected to form another part. Light A3. Inside Light A3 then returns to the back side WB of the wafer and produces the second transmitted light T2, and another internal light A4 according to the same behavior as the A1. The A4 then returns to the wafer surface in accordance with the same behavior as _A2. Produces external light R3, and another internal light A5. Therefore, we can see that a single light A0 is injected into the surface of the wafer, which can produce an infinite series of reflected light, such as R1, R2, R3, etc. A series of transmitted rays, T2, etc. In fact, the finite reflectivity and transmittance of the surfaces, coupled with the passage of individual internal light through the finite absorption of the thickness of the wafer, often results in increased light as the number of reflections increases. The intensity is quite rapidly diminished. However, if the measuring instrument collects the energy of the light rays (such as R2 and R3) in addition to the first reflection R1, the reflectance measurement is greatly affected by the transmission of light into the substrate. Similarly, the transmission measurement is also affected by the energy of multiple reflected light lines (like Τ2) in addition to the first light. The fourth figure shows the multi-reflected light effect in reflectance measurement, the plot of which is roughly the same as the third, except that the incident light is shown as a parallel light of limited size instead of the idealized single light of the third figure. . The figure shows the two ends Α and Β, representing the outer boundary of the parallel beam Η0. The overlap of the reflected beams on both surfaces of the wafer (ie HR1 and HR2) is represented by 〇vRl, as 14 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013. TSUEI.Doc 1302358 If you try to measure the reflectivity with a light detector, the light collected by the detector will not only reflect the light reflected from the front surface of the wafer, but also the light reflected from the back surface of the wafer. Such measurements do not distinguish between light reflected by the front or rear of the wafer. Similarly, the transmittance measurement is also a measurement result that is affected by multiple reflections of light within the substrate. For example, the overlapping OVT1 effect of the beams HT1 and HT2. The present invention generally relates to a method and system for emitting light onto a substrate (e.g., a half-conductor wafer) and separating the amount of light reflected from the front side of the wafer from the amount of light reflected from the back side of the wafer by a variety of methods. Once the light is separated, an accurate measurement of the reflectivity of the individual surfaces can be made. The present invention also finds that this message can be used to control at least one factor in the process chamber when the wafer is to be processed later, as detailed below. For example, referring to the fifth figure, one method of mitigating multiple reflection effects is shown. In this example, the size H0 of the incident light, as well as the angle of incidence on the substrate, are selected to separate the beams (HR1, HR2, HR3, etc.) reflected by the different surfaces. The position of the incident light beam (HR1) reflected by the front surface of the substrate and the incident light beam (HR2) position reflected by the rear surface of the substrate do not overlap. Therefore, there are multiple reflected lights separated by intervals. The light from these beams can fall to different detectors or a column of sensors. The intensity of the light reflected by the first reflection of the incident beam is only affected by the front reflectance of the substrate, while the second reflected beam is affected by the reflectivity of the front and back of the substrate, as well as the absorptivity of the substrate. In some cases, only the previous reflectivity is concerned, but the light from the individual beams can be used to more fully analyze the optical properties of the wafer. For example, collecting two reflected rays can infer information about both surfaces of the wafer and/or the absorption coefficient of the wafer. Analysis of the different components of the transmitted light ΗΓΠ, ΗΤ 2, etc., as shown in the fifth figure, can also apply the same benefits. In addition, measuring the light incident from the front or rear of the wafer, respectively, provides more information and/or reduces the degree of error in the estimate of the optical properties of the wafer. When implementing the method as shown in Figure 5, one or more of the detectors used may include any suitable means capable of measuring the intensity of a beam of a certain wavelength or range of wavelengths. Although there are many advantages to using the configuration of the fifth figure, in some embodiments it may be difficult to 15 C:\Eunice 2006\PU ^SE\PU.〇68\PU-068.〇〇13\PU-068.〇 〇 13.TSUEI.Doc 1302358 is applied to all kinds of substrates, especially for relatively thin or relatively large refractive index. In these cases, the incident beam strikes the position of the upper surface, and the position of the reflected beam from the rear surface of the substrate may be very close, so that a small incident light is required. However, a laser source can be used to provide a higher illumination intensity of a narrow beam. In another embodiment, even if the reflected beams are partially overlapped, a variety of techniques can be used to determine and distinguish the amount of light reflected from the front of the wafer, as well as the amount of light reflected from the back side of the wafer. For example, as long as the beams are only partially overlapping, the degree of overlap can be calculated geometrically. In addition, the size, shape, or angle of incidence of the incident beam can be changed to overlap, which will allow the amount of light reflected by the front to be measured. For example, changing the angle of incidence of the parallel beam 110 changes the reflectivity and the path length through the substrate. For example, in a particular embodiment, the substrate reflected light measured in one instance may include all of the reflected components R1, illusion, R3, etc., such as shown in the third figure. In the second case of changing the angle of incidence, the reflected beam components may not overlap. The same technique can also be used to analyze transmitted light. In other embodiments, instead of or in addition to manipulating the light source to separate the different light components, a light path can be configured to separate the different light components. Once different light components are separated, any light component is detected or measured by excluding other light components. For example, in one embodiment, light reflected from the upper surface of a substrate can be separated from light reflected from the bottom surface of the substrate. The amount of light reflected by the upper surface and/or the amount of light reflected by the bottom surface can be measured for determining various characteristics of the substrate. For example, the light reflected from the upper surface of the substrate and the light reflected from the lower surface each provide the characteristics of the substrate. For example, the sixth figure shows an alternative scheme for distinguishing between light reflected from the upper and lower surfaces of a wafer-like substrate. As shown in the figure, the light from a source s is advanced by the button containing the lens and the mirror. The ray R1 is emitted by 8 and then concentrated by the lens L1 to form a parallel beam with a light A2 filament. A2 continues through a mirror μ and becomes A3. L2 is a lens that focuses light, forming a ray 4 that strikes the front surface of the wafer's buttocks. Part of the light Α4 is reflected by WF to form light ARF1. ARF1 is concentrated by lens L2, which in parallel forms light ARF2. Light ARF2 is reflected by the mirror to form 16 C:\Eunice 2006\PU CASE^U-〇68\PU.〇6a -0013\PU-068-0013- TSUEI. Doc 1302358 Light ARF3. ARF3 is concentrated by lens L3, focused, Light ray ARF4 is formed to illuminate a detector D2. The second portion of light A4 is transmitted through the surface of the wafer to form internal light AT1. Part of the ATI is transmitted through the back of the wafer to form transmitted light AT2. Ατι is reflected by the back surface WB to form the internal light ray ATRB1. A portion of the ATRB1 is further reflected by the front to form the internal light ray ATRBRF, and the ATRBFR1 is then advanced to the back surface, and some is transmitted through the wafer back surface WB to form the second transmitted light ATRBRF2. ATRBRF1 (not shown) will also be reflected by wb, creating an endless series of internal light.

As discussed earlier. A portion of the light ARTB1 is transmitted through the front stomach to form a second reflected light ATRB2. If both ARF1 and ARF2 are concentrated in the optical components of the plated light measurement system and arrive at the same detection unit, the reflectivity on the back side of the wafer affects the reflectance measurement. ATRB2 is concentrated by the lens L2 to form a ray of light, and then reflected by the mirror μ to form ATRB4, and is focused by the lens L3 to ATRB5 to be irradiated to a detector D3. If D2 and D3 are separate different detectors, it is possible to distinguish the two reflected lines. In this example '>[The measured value of the detector D2 represents light that is only reflected by the front of the wafer. The measured value of the D3 can also help to analyze the characteristics of the wafer because its intensity is affected by the amount of light absorbed within the wafer, the transmittance of the WF, and the internal reflections.虿 In some of these systems, the light source (a) X cuts a number of times due to dryness. Although the tempering can be made by the limitation of the silking limit, the measurement will be weakened due to the decrease in the light intensity. The light traveling in the light towel can be analyzed at different angles of the optical axis. The seventh is the same silk system as described by the two ray diagrams emitted by s. In this example, the light 4 and the optical axis form the same emission, but both are located on opposite sides of the optical axis. Their behavior Na and Sheng represent the two rays reflected from the front of the wafer, and it can be seen that they all come to the same point of the gamma (four) plane. This is because after the optical instrument is reflected by the front of the wafer, the image surface is flat. ^ The focal length is the same as the S and u axis, (4) The object L2 _7 face 2 17 w when e fiber 叩 (10) 崎崎 (4) 6 _ The side gamma is the same as 1302358, and the focal length of the lens L3 is the same as the distance between the L3 and the detector D2. The source S is imaged on the surface of the detector by the front surface of the wafer WF. This condition ensures that the energy from the source is efficiently transmitted to a small area on the detector plane, i.e., where the detector D2 is placed. In the particular embodiment shown in the seventh diagram, the ARF and BRF rays are collected at the same point of the detector D2 because they are all emitted by the same point of the source S. In the present embodiment, the point on the light source S is located on the optical axis, but any point on the S plane is also applicable. Conversely, the rays BTRB and BTRA reflected by the back side of the wafer reach different positions on the plane of the detector D2. If necessary, they can be detected by detectors D1 and D3. As mentioned above, the light rays BTRB and BTRA reflected from the back side of the wafer can also provide useful information about the optical properties of the substrate. Thus, in some embodiments, it may only be desirable to detect light reflected from the front of the substrate, but in other embodiments, it may only be desirable to detect light reflected from the back of the substrate. Of course, in still other embodiments, the light reflected from the front of the substrate and the light reflected from the rear of the substrate can be measured separately. Figure 8 shows another optical path that can be used in conjunction with the present invention. The light path described in the eighth figure is similar to that depicted in the seventh figure. However, in this embodiment, the two rays A and B are emitted by the light source S at a different angle from the optical axis. Similarly, these rays are reflected by the front side of the wafer, and ARF and BRF reach the same point of the detector D2. However, the two rays ATRB and BTRB reflected from the back side of the wafer arrive at different positions, and only the BTRB falls on the detector D2. This roughly shows the relative impact of the different rays reflected by the front of the wafer or behind the wafer, as measured by the detector D2. The energy reflected by the wafer backside WB compared to the energy reflected by the stomach in front of the wafer may eventually be distributed over a larger area of the detector plane. Therefore, the energy reflected from the back side of the wafer contributes less to the signal measured by the detector D2 with respect to the energy reflected by the front side of the wafer. This is because the choice of optical instrument is to have the wafer surface WF imaged on the detector plane. Since the light reflected by the WB on the back side of the wafer cannot be imaged on the detector plane, its energy distribution is more diffused. The principle of separating the energy reflected from the back side of the wafer from the energy reflected from the front side of the wafer by selecting the optical conditions to optimize the energy density falling on the detector is shown in Figure 9. Curve A shows the energy density of the light reflected by the front side of the wafer on the detector plane C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358. Curve B shows the corresponding distribution of flaws reflected from the back side of the wafer. Curve c is the light that is reflected more than - times on the back side, and curve D is the sum of the light intensities of A, B, and c. If we think that the curve D falls on the X-like D/2#+XD/2, the wire indicates the month b塁 that will be detected by the side device. The curve A shows that the signal distribution is larger than the curve B or c. Much more. As shown in the sixth through eighth figures, a plurality of optical paths can be utilized and adjusted to separate the different light components, and/or otherwise control the optical signals from the front or back reflections. In addition, other techniques or optical components can be added to the optical path to promote separation of different light components. For example, you can use the money mirror in the towel to shield the light. For example, referring to the tenth figure, a specific embodiment of the optical path using the shielding member is shown. In this embodiment, the use of the shielding member (10) blocks the light B and the light closer to the optical axis. Thus, the detector D2 receives only the light reflected from the front of the semiconductor wafer. Such shielding elements that control the angular distribution of the incident surface on the wafer surface can also be placed elsewhere or as incident or reflected beams, as long as the effect is to block the light reflected by the back side of the wafer and allow the desired light to fall on the _| |上上. Controlling the incident angle distribution can also be helpful if you want the measured reflectance to match a particular angle of incidence or a range of angles of incidence. In some embodiments, it is also possible to vary the range of incident angles by means of an adjustable masking element and to measure reflected light signals at different settings. Such measurements can assist in describing the optical properties of the sample, such as, for example, describing the extent of absorption within the wafer. The optical path produced in accordance with the present invention may comprise a single shielding element (70) as shown in the tenth embodiment, or a plurality of shielding elements depending on the particular application. It is conceivable that the shielding element may be a separate component as shown in the tenth figure or may be incorporated into other optical components. For example, the limited diameter of a lens, mirror or other optical element can function as a shielding element as used herein, wherein these elements can limit the propagation of light from a particular component through the optical system to the detector. The shield element (70) can be any suitable device that can reduce any unwanted light. For example, in one embodiment, the shield member can include a device that includes a light aperture to permit only selected light to pass. However, as mentioned above, a lens can also be used,

19 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.DOC 1302358 Filter, mirror, mirror or other optical component. Another way to control the ratio of the front WF reflected energy to the back wb reflected energy is to optimize the focal length of the lenses, especially L2 and L3. The focal length ratio of these lenses controls the magnification of the intermediate imaging, which is formed in the front wp and re-imaged on the debt detector plane D2. By reducing the focal length of the lens L2 with respect to the lens L3, the magnification can be reduced. The result of this approach increases the ability to distinguish between front and back reflections. The eleventh figure illustrates this point. In this example, we can see that the light BTRB no longer falls on the detector d2. Compared with the BTRB in the eighth figure, it falls on the detector. This figure shows the resolution of the crystal. The ability to reflect light behind the circle has improved. • As shown in the tenth-figure embodiment, different light components can be separated by reducing the focal length. In the embodiment shown in the seventh and eighth figures, the front of the wafer is at the focus of L2 and the detector is at the focus of L3. The depth of field is determined by many factors, including the focal length and angle of illumination. If the condition is set such that the light emitted by lens 12 and lens u from a predetermined narrow range near the surface stomach is only imaged at the detector plane, then the depth of field is confined to the area near the surface of the wafer. By ensuring that the wafer backside WB falls outside of the depth of field, the light reflected from the front and back of the wafer is separated because most of the radiation reaching the detector is reflected by the upper surface of the wafer. In some embodiments, the incident beam can utilize a relatively large range of angles of incidence. This φ condition also represents the range of incident angles at which the heated radiation is incident on the wafer in the process chamber, providing the second advantage of this measurement. In this way, an optical model can be used to estimate how the radiation is radiated onto the wafer in the process chamber, such as using trace tracking software. A modest understanding of the range of incident angles of the heated radiation can then be used to match the illumination conditions used in the measurement system to the illumination conditions implemented in the process chamber. In some instances, the reflected light can be collected from a relatively small area on the wafer surface to avoid collecting light from the back of the wafer and being reflected toward the front. The size of the area of the reflected light that is collected depends on the optical instrument used, the size of the device, and any apertures, filters, or other scales that are included. The optimum size of the analyzed area is partly based on the sample thickness 'because if it is a sample (four), if the light is selected from the wafer - the bulk area is even more difficult 20 C:\Eunice 2006\PU CASE\PU-068\PU- 068-0013\PU-068-0013-TSUEI.Doc 1302358 ^Open the front and back reflected rays. However, "other factors need to be considered. For example, if the wafer has a surface with a pattern, the reflectance in the area where the reflected light is collected may vary. It may be better to collect the average characteristics of the area at this time.

For example, in one embodiment, the area of the wafer on which the reflected light is to be collected may be relatively small in size and may be designed to fit the area observed by a temperature measuring device such as a pyrometer. In other words, the field of view of the pyrometer can be matched to the area of the area where the reflected light is collected. However, in other embodiments, the light may be collected from the relatively large area of the crystal BU. 2 Collecting light over a large area of lx can help optimize the heating device and the substrate. In other words, if the substrate is heated by an energy beam, the area describing the optical properties is matched by the energy-reflecting domain. To optimize the energy balance of the Wei beam, the description of the two characteristics' may be particularly useful. This is particularly useful when the wavelength of the emitted energy is greater than about i _ , where the semiconductor substrate is typically semi-permeable. For example, such lasers include diode lasers, her stone picking lasers, fiber lasers, co and C0r lasers. In a self-contained embodiment, the light source can be scanned across the substrate. The intensity of the light can then be reflected, which can be achieved by moving the light source, moving the light path, and/or moving the wafer itself. For example, t can be used to collect information on any of the specific locations on the substrate. Green, which is the average of the entire surface of the wafer. In the -frequency real_, you can use the "micro-thief reading" to provide a short-term lens. This lens can be placed anywhere in the path of the light' as shown in the eleventh figure. If placed in a lens, it can be focused on the surface of the wafer. This series of mirrors can have a short focal length. Therefore, 'this type of lens provides a convenient way to allow the optical system to illuminate the angle of the human lens' and collect light at a short focal length, especially the magnification of the wafer surface. (9) If the 'scale is more than 10 times, it is more than 50 times. 'However, there are many alternatives that can be considered, including the collection of only the required areas, and most of the collected Komatsu shots are only reflected by the brother-surface. The optical axis of the measurement system does not need to be perpendicular to the wafer surface. The analysis of the light distribution on the plane of the controller can be detected by scanning the detection on the plane, or by making the size of the detector (4), or the variable or on the plane of the controller ^ 21 C:\Eunice 2006 \PU CASE\PU-068\PU-068-0013\PU-068-0013 TSUEI.Doc 1302358 Scanning a light hole, or by using an image detector like a charge coupled device (CCD) camera, or a detective Array of detectors. Information about the spatial distribution of light on the detector plane can be used for fine measurements. For example, if the thickness of the substrate is known or measurable, it is then possible to interpret the density distribution pattern with respect to the composition reflected in front of or behind the wafer. An algorithm can be used to segment the measured distribution out of the components reflected by the front, rather than the components produced by multiple reflections in the substrate. Thus, as shown in Figure 9, the component reflected from the back side can be mathematically subtracted from the total signal to obtain a more accurate estimate of the signal reflected from the front of the wafer. In the fifth to eighth figures, and the tenth to eleventhth views, the focus is mainly on distinguishing the amount of light reflected from the front side of the wafer from the reflected light on the reverse side. As mentioned above, there will actually be a series of endless reflected rays, and the twelfth figure shows how a single ray A from the source 8 is incident on the optical system as described in the ij article. In this example, the figure shows the path of the light ray R1 reflected by the front surface WF of the wafer, and the light ray R2 reflected by the back surface WB, and also the 3, R4 and R5, which are respectively two or three times on the back side. And four reflections. In principle, this sequence will have an infinite amount of light, but as mentioned earlier, the intensity of the lost light is reduced rather quickly. Light R R R ' R3, R4 and R5 reach detectors D2, D3, m, ^5 and D6, respectively. It is possible to measure the individual intensities of these rays in order to better describe the characteristics of the wafer. For example, to measure rays Rh R2, R3, R4, and R5, the source s can [laser so as to produce sufficient intensity to be measured. ―For a source that emits a range of angles, there will be a lot of light falling on each side element, resulting in a sequence of curves like B and C in the ninth figure. The analysis of the optical density profile helps to identify the components as described above, and this method can be used in conjunction with the use of optical apertures or other optical components that can be injected into the scheme, which is also dependent on the sequence reflections of the detector. Light composition. Above, the principle of analyzing reflected light can also be used to analyze the light that penetrates the substrate. For example, the specific complement of the thirteenth _ includes two rays A and B, and the light source s is incident on the light path, and the lenses L1 and L2 function as described above to image on the wafer surface, and the lens Μ and L5 will The s image on the WF is again imaged on the detector plane. This design ensures that the transmitted light is not reflected by the back side of the wafer, in the plane TD2 22 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI. Doc 1302358 Difficulty. Light rays that are reflected at least once through the back side of the wafer, such as ATRBRp and bTRBRF, arrive at different locations on the detector plane. Thus, in a manner similar to the reflected light, the contribution of the multiple reflected light to the signal measured by the TD2 can be reduced. . Figure 14 shows the path of a single ray A that is transmitted through the optical system and the crystal to produce a plurality of transmitted rays in a manner similar to the multiple reflected rays depicted in Figure 12. If desired, such light can also be analyzed by the side-H array side and the spatial distribution of the transmitted light is analyzed in a similar manner as discussed for the spatial distribution of the reflected light.

^ It is conceivable that although the illustrations herein show the use of two lenses between the source and the wafer, 'a variety of alternative optical paths that accommodate multiple or fewer lenses can also satisfy the general principle, j is a mirror ( Some concave lenses, such as concave mirrors or other optical components. An optical instrument that controls the polarization of incident or reflected light if f is required may also be included in the device. Compensator H can include polarizers and dimmers, such as quartz, half-wave or full-wave, other optical components that produce the desired polarization (including planar polarization, elliptical or circular polarization). Once the polarization state is configured and/or the angle of incidence is configured, it is conceivable that any known measurement = arrangement is inconsistent. For the straight person shooting straight line (that is, the incidence of incident light) The angle is less than about 25 degrees), and the effect is usually quite small. In the non-vertical input = example ', the human-radiated W-plane or p-plane polarization can help the side. The two kinds of Z: can be measured comprehensively. Optical characteristics. Ellipse measurements can also be performed on multiple days of the same day, such as during the low-temperature description of the substrate during its characteristics. It can be performed at any desired wavelength, or by connecting a suitable spectrum device (like ★士5', His wavelength selection filter ^ cross-spectrum implementation. Measurement of comprehensive characteristics;: == source 'and use TM long response coffee - bulb: Xenon fox light, light-emitting diode: laser, or heat _ plant ° , hot rods or other components that emit heat and light radiation.) In some embodiments 23 C.\Eumce 2〇〇6\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc In 1302358, the source of the radiant source of the luminaire is set to the spectrum of the energy used in the process equipment to represent = to such a square di - is _ _ _ The child and where it had binding member, energy simulation system used in the process equipment. Filtered

In the specific implementation, the light source may comprise a supercontinuum source. A shank light source has some advantages for quickly measuring the filaments of the wafer, because the two d-special Kuanxian source (like - crane wire, halogen lamp or - LED) is also bright. This can help with very fast measurements because the energy delivered during a fixed time period is much greater, allowing the reflected side to achieve a higher signal level. The filaments are at a relatively high level of interference, interfering with the accurate measurement of the characteristics of the wafer, and the supercontinuum source of light will also be unimportant. This can be particularly advantageous for wafers in the process chamber, especially if the wafer is hot or has a heater, a lamp, or an electron emission. The extremely bright source of radiation also allows the use of small, fine =' = large amounts of radiation from a small launch area, and can be easily combined with ten, sense, material and other components that can be introduced into process equipment and process chambers. Super-light 2 can produce a fairly flat spectrum, that is to say it has a wide wavelength ° - the advantage is that it can be conveniently covered by the silk / sub-simplified spectrum measurements (like reflectivity) And the interpretation of the transmission spectrum). Supercontinuous spectral light is produced by exposing a nonlinear medium to high-energy light. For example, 'a high-energy radiation thief from Wei can be used—water battery (5) (10), and then produced in the county where the light is emitted. - Effective methods include the use of photonic crystal fibers or tapered f-wear. The pro-continuation can produce a squama wavelength spectroscopy, which helps to describe the semiconductor 曰曰 round special! · For example, the 'k perK white super-continuous light source generated by K0HERASA/S (B oil (10)d, Denmark) A spectrum of radiation in the spectrum of the 胍 胍 and 2 〇〇〇 = = greater than 4 mW / nm. Such a light source is used to measure the wafer transmittance or reflection of the process (4). The amount produced can be used to bind the Crystal® temperature. In addition, the source can be used to verify its _ quantity, or money or combined with the wavelength of the shouting component. The member 7L can select the wavelength of light transmitted to the wafer, or the wavelength of light that is refracted, scattered, transmitted, or shattered by the wafer. Measurements may include reflectance and transmittance as described above, but may also include a 24 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 It is modulated when it arrives at the wafer and causes thermal or electronic modulation of the properties of the wafer. This modulation can be detected and used to obtain information about the wafer. In addition, such light sources can even be used in thermal processes, especially where there are coating patterns on the wafer. In this case, the use of broadband source heating reduces the amount of difference in energy absorbed by different pattern blocks, providing a more uniform process.

If you use thermal energy, it is best to set the temperature to the representative temperature below the actual process conditions. If necessary, the latter can be set to match the wafer temperature. ^ mouth, if the key process steps are carried out at lGGGt: the ship can be set at this temperature. In the case where a wide-band light source can be used, it is preferable that the system-broadband generator be a thermopile, a light-emitting meter or a thermoelectric reading to detect reflected radiation. Differences in the conditions of the process chamber and the axis of the meter can be compensated for by the calibration procedure. The recording device uses _laser to heat the wafer and may have to optically measure at the same wavelength. Lai Lei shots are injected into the wafer with a special human angle of incidence and polarization, and these considerations may have to be included in the measurement. It is conceivable that the front and back measurements of the substrate can be used to provide a more complete understanding of the optical properties of the substrate. As noted above, in a particular embodiment, measurements can also be made at a number of locations on the wafer surface. This message can provide a spatial distribution map of the crystal emission characteristics. Measuring at multiple locations on the surface of the wafer may be particularly useful for improving the uniformity of the process. For example, regarding the optical or thermal energy on the surface of the wafer, it can be supplied to the rider or control system. In a particular embodiment, this is used by the model-based control 11 in order to optimize how the wafers are different: the heating condition. For example, the distribution of the characteristics can be provided to the model-based control system t to the optimal deposition of the f-source of the lamp, and (10) the method of optimizing the temperature uniformity 1 can also be used to control the addition of the substrate by the thief in other ways. Energy coupling variation when applying one or more laser beams. The message of the variation of the optical characteristics of the ―, ί helps to correct the temperature measured by the 上 on the wafer. Information about wafer thickness or doping variation can also help: for the sake of purpose. For the same reason, the overlay or pattern on the wafer can also be used accordingly. 25 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 In the green port example, it may be useful to measure in at least two configurations, one in the bean /, collecting the reflected light from the illuminated surface, the other optical configuration is to let the light reflected on both sides of the wafer (such as the depth of field in the third picture accommodates n · by changing the optical path independently Steps such as oblique riding without a light hole, or by measuring the reflectivity of both sides of the wafer and the measurement of transmittance, the characteristics are described. Transmittance at room temperature =, determine whether the wafer is heavily doped This can be caused by a wavelength greater than f ΐΐΓΐ. The absorption coefficient is much higher than the value of the lightly doped material. This message is followed by a new temperature 驴 面 进 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是 是_. Especially in 曰 π cases t & can be used to distinguish between lightly doped wafers and heavy doping, or provide extensive information on doping properties. At more than one temperature: Line 2' can also be used to determine other characteristics of the wafer. For example, _ of different surface reflectivity can be displayed The temperature of the sulphur rate. The temperature is estimated to be correct for the reflectance, transmittance, or absorptivity. For example, the surface reflectance at the center of the heart (T2) can be obtained by other temperatures. The measured values are extrapolated, such as T1 and T3. The value is described at more than one temperature and then the value of the third, field is extrapolated'. This concept can also help to estimate the temperature dependence of the substrate absorption coefficient. "These measurements" can be combined with the light transmittance of the substrate, for example by illuminating the ffiH 额外 additional self-domain wei, such as the electronic space made by the semiconductor towel. It will be reflected by the reflection of the money _ component _ should be modulated, 14 components are susceptible to the thickness of the light traveling through the plate. This method can improve the quasi-heterogeneous. We can also by applying its teacher's touch (including electrons) To modulate the light transmission, this electronic ride can be viewed by the mechanical modulation of p ^ from the P-radiation emitter. It is also possible to obtain additional information by deliberately looking at the wafer temperature and observing the change in the property. 26 C fine "Gui Na Na 0 turn 〇 6 purchase (four) Gamma plus 1302358 As mentioned above, the wafer surface can scatter radiation due to variations in pattern or surface layout. If so, the intensity of the reflected or transmitted light beam can be affected by the scattering pattern. And the degree of miscellaneous, that is, to verify whether the transmittance measured by the two-sided shot is the same. If the transmittance varies depending on the surface of the irradiated surface, then at least the surface of the wafer may scatter light, which is scattered. Direction cannot be accurately collected

the amount. Therefore, the 'asymmetric transmittance measurement value may be the symptom of light scattering. The neon storage (four) miscellaneous, which can be improved, can improve the optical properties (4). In order to propose an improved scale of the pyrometer wavelength, it is particularly obvious that if the light scattering effect is particularly obvious, it may be necessary to measure the wavelength of the wafer! The wavelength is a function of the spectrum of the energy emitted by the energy source. Sex 2 takes into account the considerations in this regard. The optical energy can then be used to transmit: frequency ==. For example, if the property of interest is - quantity_, Ί曰ΙιΧλ), then the characteristic obtained by the integral is ρ(τ), where

F(T)

JW)fa,!T)dA !h(A,r)dA Energy. ^!^ _ Wavelength is fine, and the energy of most energy sources including the heat source is reflectivity, absorptivity, transmittance, and material. It is especially important to measure and integrate the emissivity, ie the fractional rate. The same applies to the measurement of the radiation spectrum. You u. The rate is right, the weighted spectrum is the black body of the temperature to be measured as follows σ, ~ the emissivity Stot(T) can be calculated from the spectral emissivity ε(λ, Τ), the equation € ί〇ί (T) @ J^^ , TXl, T)dZf%AArm 27 C:\Eunice 2006\PU CASE\PU-068\PU-068-mi3\PU-〇6a.〇〇13 TiUEi 1302358 where Wbb(X,T) is the description - black body - Recording, ray-type. Most of this, , and Pakistan can be: The α of the integral emissivity and the _shot_ emitted by the body at the temperature of the tau. Assisting in the determination of wafers in any of the accepted methods: the P alternative method with the appropriate characteristics of the rider may require careful ordering of the illumination frequency = private measurement. The latter side may have a faster and simpler detection spectrum in some cases, but as described above, in a specific embodiment, to provide a panel for the optical characteristics of the substrate; ^Measurement in both the face and the back', the crystal __ 卩 卩 α α (4) ^ Xie example For example, the fifteenth figure shows the ability of the arrangement surface to enter the light. The ray of the radiation source _ circled on both sides is collected by the DRW, that is, the __ shot is emitted on both sides (w). Reflected by w can include PH, rmAM, & emission collection and sensing configurations. DRWF points if needed. The same as ^1^^=^? is only the light reflected from the singer into the collection and sensing configuration of the S-day sniper. Radiation source (TM}; ^ ^ ^ « to the Han 々 configuration. Right need, DRwb can include optical instruments to limit the light component reflected by the sensing 曰 ^ 聪. Similarly, dtwb is collected transmission = ^ The combination of these measurement sub-systems allows the measurement of the σ surface to provide a more comprehensive description of the characteristics of the wafer. The two sides of the crystal® are the force of the surface-emitting light. The source SWp illuminates the front of the wafer (medical). The SWF ray reflected by the chat is collected by D1, ie a radiation collection and sensing configuration. If necessary, m I includes an optical instrument to limit the sensed light shot to only the light component reflected by w. Similarly, D2S collects the collection and sensing configuration of the SWF-rays transmitted through the wafer. Radiation source SWBA? The back side of the wafer (WB). The arrangement of the optical instrument in this arrangement is to allow the Swb light reflected by the article to be collected by D2. If necessary, D2 may include an optical instrument to limit the sensed radiation only by WB. The reflected light component. The same D1 can collect 28 C:\Eunice 2006\PU CASE\PUO68 \PU-068-0013\PUO68-0013-TSUEI.doc 1302358

$ shot through the sWB radiation of the wafer. The measurements that can be implemented in this embodiment are the same as those in the fifteenth diagram, but this implementation of the lake to fewer optical components and _ devices would therefore be less expensive and simpler. With the sixteenth scheme, it is also easier to carry out the measurement in the same space on the wafer. If there is a need to 'measure the light 仃 from different __ by sw and SWB to avoid the radiation from the Swf when the radiation from the SWB is sensed, the radiation from the Swf can also be = or can be used by using swf or The modulation of the radiation output of at least one of the SWBs, the second measurement of the value 'in order to provide a time-varying characteristic of at least one signal', the good area distinguishes that the signal is caused by the specific filament (t). For example, the SWF turns to adjust the U degree by a known frequency and the signals from m and D2 can pass through to track 2 of the frequency component. Other methods of separating the signals may include differences in light output from the SWF and SWB, 疋 wavelength or polarization state. It is also possible to describe the characteristics of the light collection and the changeable sensing configurations D1 and D2 so as to be more suitable for light shot measurements from sw or qing. For example, when D1 is collecting; k wafer is reflected from swi^ light, the characteristic of D1 can be set to optimize the collection of reflected light 'g 1 can be set to optimize when collecting light transmitted by SWB Transmitted light set. Of course, this function is shown if you want to optimize its miscellaneous. ^15 and 16 shows the characteristics of the light collection and sensing configuration that can be optimized for measuring wafer characteristics, including adjusting or selecting optical components (such as lenses, first, sub and side) Or the shape or shape. They can also include adjustment filtering or polarization. 2 We can include electronic or bit filters, amplifier settings, and signal processing. The Korean source—and S· can also include optical and electronic components that can help optimize reflection or ride component _ quantities, and these components can (4) be used to measure specific characteristics. Miscellaneous fifteenth and sixteenth fiscal _ structs show light = 曰 ^ _ straight line - angle shooter ' If necessary, it can be vertical or nearly vertical itil. Scales, which may include split wire to allow for gauge collection and sensing to configure a sample to measure crystallographic or transmitted radiation without completely blocking the path of human shots (4). / Γ 多 多 ’ ’ 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送 传送A feature of this type of optical instrument is that its focusing characteristics do not vary with wavelength. So in the sample inspection by the same area on the wafer

29 C:\Eunlce 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.DOC 1302358 Covers a wide range of wavelengths for measurement (continuous or simultaneous). By using the reflective optical age, (4) at achromatic error is caused by =: the middle measurement is performed at more than one wafer position. This amount is easily extended to use her project on the back of the wafer to measure the "characteristics" of multiple locations. The machine can be moved mechanically—measurement configuration to measure different positions ί=Preparation can be reduced-stationary crystal _, or wire measurement _beam can sweep the wafer and analyze the reflected and transmitted radiation beam . This beam may contain light that is reflected by a crystal or a sub-region (like a ""linear region). This analysis corresponds to the collection and sensing configuration implementation of the light beam, or to the collection and sensing of the (four) beam. Or the 'beam' can be collected by the optical system and transmitted to the imaging device, a camera. Practical Applications... Once we have measured the surface reflectance (and the transmittance of the wafer if needed), we can use this information to estimate the thermal radiation characteristics of the wafer during the wafer process. For a process with a wafer temperature T greater than 7 °C, there is a very simple way to use it. In this case, a germanium wafer having a thickness greater than 600 μm can be considered opaque for all wavelengths noted. If so, for most considerations, the refractive index of any known wavelength and the wafer emissivity can be linked by Kirchhoff's law. A more detailed approach can incorporate information on the wafer temperature to give an improved estimate of the radiation characteristics. For example, if a pyrometer reading is known, the temperature thus estimated can be used to improve the estimation of wafer characteristics in a number of ways. One of the methods is to estimate the absorption coefficient of the wafer from a model, and then combine this characteristic with the optical characteristics measured at room temperature to propose a more accurate estimation of the high temperature characteristics of the wafer. The number of South Temperatures can also be used to provide a better estimate of the loss of the far-wafer. this

30 C:\Eunice 20〇6\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.DOC 1302358 Political loss and ‘shooting characteristics and temperature effects. Combined with the estimation of these two characteristics, an improved estimate of the heat loss on the wafer can be obtained, and accordingly, the yen can be adjusted by adjusting the power of the lamp moderately and the temperature control is uniform with the temperature. For example, the method detailed below is to collect the light path collected in the indication. In the specific embodiment, the present invention relates to the correction-radiation sensing element 绿 绿 绿 绿 绿 绿The method has a transmittance of less than 〇 at a high temperature. "In the example, the optical path of the present invention is used to collect the φ of the two circles reflected from the surface of the wafer. The measurement is that the wafer is in the opposite direction. It is carried out at a lower temperature, for example, such as a chamber, and is carried out in substantially the same wavelength range as the operation of the pyrometer.

It is the method of revealing that the reflectivity in front of the wafer is lost. This reflectance ^ reduces the reflectance of the wafer at high temperatures. More specifically, since the transmittance with temperature is reduced, the reflectance ratio at the high temperature is substantially the ratio of the reflectance to the front of the wafer, and the emissivity of the wafer can be derived. For example, the emissivity is usually equal to 1 accuracy. ~ The required emissivity is then used to correct the pyrometer reading and to improve the temperature control. The method of the present invention can also be used to correct the pyrometer. The internal transmittance of the wire plate is greater than that of the wire plate. Ιι. In this embodiment, the amount of light, which can be used for the day ft, can be reflected by (4) - the reflection of the wire is relatively low temperature and is operated by a pyrometer. The optical configuration of the present invention can also be used to collect light reflected from the opposite side of the wafer, as described above. From this message, two == reflectivity and reflectivity behind the wafer can be judged. The wafer's transmittance can then be measured as measured by the _ model or by other methods. Clear break = the absorbance can be estimated from the reflectance, usually the same as the emissivity at high temperatures. The sum is ((4) ϊ and the receipt is at 1 minus the radiance. The available absorption rate optimizes the energy absorption or the power setting of one or more heating elements in the process chamber. [Methods are particularly suitable for systems like those depicted in Figure i. The wafer (10) is made of 31 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-06a-Q〇l3-TSUEI.Doc 1302358 thermal element (22) from the wafer Side heating. However, if both sides of the wafer are added by different heating elements, this method can be repeated on the opposite surface of the wafer. As such, heating the first heating element on the top surface of the wafer can be controlled separately from the second heating element that heats the bottom surface of the wafer. - The method disclosed in the present invention can also be used to optimize the setting of the heating element at a lower temperature 'where the internal transmittance of the substrate is greater than 〇·· In this embodiment, the optical instrument configuration used can collect and determine C The amount of light reflected by the surface, at this time the wafer is at room temperature σ or 接, to / dish and the wavelength range is substantially the same as the wavelength range used by the heating element to heat the wafer. Then, the wafer is subjected to the same measurement determination on the other side with similar conditions.

From this message, the reflectance in front of the substrate and the reflectance on the back side can be measured. The transmittance of the wafer at high temperatures is then pushed or pushed out. By the assumption that the absorption rate is basically different, the absorption rate can be estimated by measuring the transmittance. Therefore, ΐ ifί subtracts the transmittance minus the reflectivity of the wafer side. The absorption rate can then be used to minimize the heating element energy output of the read energy output and/or the crystal form. The second heating element heats the wafer to optimize the heating alone. If the first heating element heats the wafer, the method element can be repeated on the wafer opposite to the other side. Heating; ==!== The information obtained by the disclosed method can be used for control, and the learning characteristics can be as shown in the tenth figure. The plane of the internal test ^, or - the mechanical arm = t in any fit in the charm (9) to the middle of the above measurement can be closely followed by the second 丨. The picture shows a complete wafer through the 糸 ^ MW00) 〇 ^

Round conveyor (110). In another part of the J, the system further includes an R to the wafer optical processing chamber during the process, including wafers in the card)

C:\Eunice 2006\PU CAS^-068\PU-068-0013\PU-068-0013.TSUEI.Doc 32 1302358 The optical properties of at least one Crystal® are determined by the method. Once the wafer is special (10). The crystal optical characteristics measured in the process chamber (2〇〇) are transferred to the thermal process chamber by a wafer transfer device (11〇), which can then be used to control the second Two process variables or system components. For example, this message can be used to control, the ship's m (four) power controller, or a pyrometer that can be used to calibrate or otherwise control the temperature used to measure the daily solids.

(10) Re-/Return to Know the First Diagram The thermal process system (10) may further include a system controller such as a microprocessor. The configuration of the controller (50) is designed to accept from the light G) (4) squeaking, representing the hoof of the different positions. The temperature of the wafer (10) can be calculated based on the optical characteristics sensed by the photodetector (), coupled with the optical characteristics measured in the wafer optical processing chamber (200). The system controller (5〇) as shown in Figure i can also communicate with the heating element power controller (9). Similarly, based on the optical characteristics of the wafer measured outside the process chamber, the control (10) can selectively increase or decrease the supply of the heating element 7L to optimize the heat energy radiated by the heating element and be absorbed by the wafer. The heat. However, in addition to controlling the light intensity, it is conceivable that the power controller (9) is equipped with a system controller (10) that can be used to control the heating element in other ways. For example, the system controller (50) can also be configured to change by The light dose emitted by the luminaire (24) causes different parts of the wafer to be subjected to different amounts of light. The angle of the human being when the radiation is in contact with the Japanese yen (10), and the wavelength of the radiation can also be selectively Control of the Invention Before discussing more clearly the various methods in accordance with the present invention, it will be useful to first discuss how light passes over the wafer and how the light intensity changes depending on its path of travel when the wafer contains front and back coatings. For example, referring to Figure 18, a substrate or wafer is shown (one sentence includes a front coating (8〇) and a back coating (82). An incident energy (84) and a wafer (14) are illustrated. Front Coating (8〇) Contact. The general wafer (14) shown in Figure 18 has a number of properties to be considered when implemented in accordance with the methods disclosed herein. For example, two surfaces of a wafer It can have different refractive indices and transmittances. In addition, the reflectivity of the surface may be dependent on the radiation from the outside of the wafer. Table 33 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-〇6a.〇〇i3.TSUEl Dot 1302358 The second S is different from the surface. The surface area may include various thin surfaces. The substrate material covering the wafer base is formed. If a wafer is semi-transparent: t: multiple reflections of different energy beams traveling, will affect Observed by the outside of the wafer = and its transmittance 匕 all optical properties may be wavelength in and

In the discussion of Xuanxia, 'Tt is the transmittance above the wafer' Tb is the transmittance below the wafer', and the twinned BU wire is reflected by the surface of the outer surface of the wafer. The reflectivity of the radiation thereon, and is the reflectivity of the underside of the wafer for the light shot incident on the substrate. In general, if the incident light shot is not a vertical shot, all of the above characteristics will be a function of the human angle of incidence and the plane of polarization. In general, the material of the wafer body has an absorption coefficient σ(λ, τ) which is a function of the wavelength of the radiation and the temperature Τ. The intensity experienced by the light passing through the substrate is weakened, and the following magnitude can be used to describe the field a = exp(-a(A, I>i/e〇sg), where (1) where d is the thickness of the substrate and $ is the internal angle of travel. The angle is the angle between the direction of the light and the perpendicular to the surface of the wafer. The ra used here refers to the internal transmittance. In the nineteenth figure, we can find that the intensity of the reflected light R1 is only reflected by the front of the wafer (WF). Rtv affects, so if the intensity of incident light is 〗, then the intensity of ray R1 is RtvI. The light transmitted into the substrate has a strength TtI at the point where it passes through the front region into the wafer body. When ray A1 passes through the substrate, due to The energy is absorbed and loses strength. Therefore, it has a TtI intensity only at the point of reaching the back surface region (WB). The portion reflected on the back surface forms the internal light A2 having the intensity aTtRbsI, and the transmitted portion forms the light Tb with the intensity aTtTbI. When the reflected light A2 reaches the front, it loses more strength due to absorption in the substrate, and now the intensity is a2TtRbsI. A2 is reflected in the front portion to form the light A3 having an initial intensity of a TtRbsI, However, it is transmitted back through the front part to form light 34 \Cunfce 2006\PU Ca^PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358, with intensity as a function. When light A3 The intensity of the back surface is a3TtRbsRJ. The reflected portion of A3 forms light A4, the initial intensity is, and the transmission is transmitted through the back ^^^^3TtRbsRtsTJ〇 and a TtRbs RtsI. The light reflected from the front forms the light with the intensity of Α5 2 and the transmission _ part of the formation intensity is a4Tt2Rbs2RtsI ray R3. As such, = easy to see the multiple reflection occurs, the transmission of the transmitted light from the substrate, the mother channel is more attenuated relative to its former term. Occurs as a result of crossing the substrate twice and being reflected upwards, so the intensity of each ray is attenuated by the front-beam beam. You sigh a RtsRbs. The total intensity transmitted by the wafer can add all, T2, etc. In total, the order of the form is +ατΆ{αχ^aTtTb_bsp. (2) and can be simplified to the following formula. The total intensity reflected by the wafer in the same way can be used to form all the components to form the following number of gorges ( 4) and can be simplified into the following calculation / A + | ten ten) 2 + + m ======= material. Total penetration C:\Eunice 2006\PU CASE\PU.〇6a\PU.〇68-0〇l3\PU.〇68.〇〇13 -TSUEI.Doc 35 1302358 "Xv m is the same as the front of the wafer" The reflectance is obtained by the following formula

The Rm problem can be influenced by the angle of incidence and the plane of polarization. This argument. At each calving, X, Y, and P-polarization and S-polarization are taken into consideration to determine the transmittance. For the reflectance and awf corresponding to the reflectance of the wavelength of the month, the reflection ratio t and the transmittance of ^1\, the following formula can be used to calculate, the angle and the polarization_crystal front emission rate or the absorption rate (8) =1 AWF ^ ^ and Equations 6, 7 and 8 together can be inferred (9) 222 AWF = 1 - ^ 1-heart A ' and the absorption coefficient and thickness of the substrate, etc., quotient: ΐ:: reflectance of the road 1 Transmittance, the absorption of the sheet metal plate, k Tianbei; bucket 'can use this formula to calculate the emissivity in front of the wafer or absorb a certain characteristic. n sub-independence contributes to the total transmission or reflection intensity of the wafer. Different anti-t or, shoot light. As a result, there is a loss of information that may help describe the characteristics of the wafer. For example, by the intensity of the secret light R1, the quantity ‘ can be directly obtained, and it is not necessary to know a, Tt, Rbs or Rts like ^罝RWF. _, the quantity aTtTb can be determined by selectively measuring the intensity of the light 1 without having to know the heart or Rts as when measuring s. In addition, more information can be obtained by measuring the intensity of other rays in a selective manner. E.g,

36 C:\Eunice 2006\PU CAS£\PU-068\PU-068-0013\PU-068-0013-TSUEI.DOC 1302358 fTo determine the internal transmittance a, then the absorption coefficient can be calculated using Equation 1. The absorption system is calculated by rearranging Equation 1 to obtain cos0f - d Ιϋβ, (10) and ei is ο at normal incidence, and the above formula is more simplified into 〇^(λ, Τ) hia ~ (11) ❿ - if radiation is not Injected at a normal incidence angle, the angle Θ can be calculated by Snell's law. The refractive index of most semiconductor materials is quite high (greater than 3), so the approximation of θ is usually caused by an error of less than 7% for the absorption coefficient, and if the radiation of a round is irradiated The angle of incidence is less than 30. , then the error is usually less than 2%. The value a2Tt2Rbs can be calculated by selectively measuring the intensity of the light R2. This is especially helpful because 'depending on the internal transmittance, let us use the composition of the reflected light to estimate the amount of optical absorption in the substrate. The amount of optical absorption in Reiner is usually determined by performing a transmittance measurement. In some cases, it may be difficult to apply due to its geometric shape, mechanical or other limitations. In this case, the _reflection measurement can be used to achieve the target. Although Rwf is also affected by the internal P transmittance, it may be difficult to calculate the internal transmittance using the directly measured Rwf. The intensity of the =j ray R1 is much larger than the ray R2I, so the fine measurement is the main surface reflection & this value does not affect the internal transmittance. Therefore, it may be awkward to ignore the contribution of light, R3, etc. to reflectivity. This makes it easier to have errors in reading transmittance based on Rwf. In contrast, the 112 intensity is directly affected by the internal transmittance, so its measurements give a more accurate estimate. The above-mentioned characteristics can also be determined by the plastic film 1 μ which is also affected by the top surface transmittance of the wafer and the shadow 112 of the back surface reflectance. Since the reflectance RWF of 112, R3, etc. is also susceptible to the above characteristics. However, it is also subject to ~ and ts "' so it may be more difficult to use the measured values of Rwf to extrapolate Tt or '. For any other individual choice of light contribution _ amount, in many different cases also C:\Eunice 2006\PU CASE \PU-068\PU-068-0013\PU-068-0013-TSU£I.Doc 37 1302358=Light T2 is susceptible to a, Tt, Tb', and & values of the radiance of this ray The 篁 value provides a reflection ratio for the light shot from the inside of the substrate. The ratio of the line and the degree can also be helpful. For example, the intensity of the light 2 is greater than the intensity ratio of the above τι, or R3. The back of the value and the front of the value describe the characteristics of the wafer. It is helpful to arrange the separation of the transmitted light components by the two sides of the crystal w. For example, the light A0 can be shot at the younger The figure is marked as WF_1, or can be shot on the side. The T2 line is the same as the Xiaodu shot and has the same partial fiber, and the transmitted light components T1 and r are also the same. If you review the items in the formula 2 You can see =^ to replace subscript b with subscript 1 for each item (each corresponding to a transmitted light rft. In the mathematical expression, the subscript is replaced, in fact The surface of the ride is transmitted by the 曰-shooting condition, which can be used to check the measuring device (the mode of the two sides of the right wafer with the two sides of the right wafer that will cause light scattering or the light-transmitting line will be More complicated, and the intensity measurement may be = side (four). If the wafer body will scatter light, it may encounter the same shape, although the _ is that - the singularity component should be _, _ does not apply == round 3 eve The facet provides the reflectivity R WB of the new news about the optical properties of the wafer. R Qing is like Equation 7, but r*wb is as follows

Rm a (12) where Rbv is the crystal back reflectance of the wafer on which the wafer is shot. In the same way, multiple reflected rays from the front (reported as R1WF, R2WF, R3 38 C:\Eunice 2006\PU CASE\PU-068\PU-〇68-( °°13\Ρυ.〇68. ° 〇^-TSUEI.Doc 1302358, etc.) is not necessarily the same as the corresponding back-illuminated reflected light (denoted as Rl^, ;^·, R3WB, etc.). If we examine the number of stages in Equation 4, The origin of the asymmetry is clearly visible, because if we replace the subscript b with the subscript t, the items (each corresponding to the intensity of a reflected light) will change. This asymmetry makes the measured value of the rear illumination Additional information regarding the optical properties of the wafer can be provided. More specifically, we can find that the first reflected light RlwB can provide a back reflectance measurement for light shots from outside the wafer. In addition, the second reflected light 112 The measured value of _ provides a measure of the backside reflectance Rbv for radiation emitted from the wafer. As discussed above, it is apparent that the reflected and/or transmitted light components are measured by illumination from the front and back of the wafer, It provides a lot of useful information about the optical properties of wafers. It is also known that it is possible to extract information about the optical properties of the absorbance or emissivity measurements on one or both sides of the wafer. These values are usually also different on both sides of the wafer. Emissivity sWF and Absorption Awf in front of the wafer It can be derived from Equation 8, and then the equation = where the emissivity of the back side of the wafer and its absorption rate are as follows (13).

R (14) ^ These two values require special instruments and are still feasible. For example, any known second rate can be derived by measuring the thermal radiation of the wafer. By comparing the heat yield of the wafer, it can be used to measure SwF and ελνΒ. It is related to the absorbance of any known wavelength. : The increase in wafer temperature can be used to determine Awf and Awb 0 39 C:\Eonfce 2006\PU CASE\PU-06B\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 Measured values Obtain information about wafer characteristics The measurement methods described in this specification can be used for a variety of purposes, including describing the characteristics of the type of wafer being processed. For example, an optimal process approach may require information about the substrate material, its doping, the thickness of the wafer, and the reflectivity and transmittance of the wafer over a known wavelength range. Such information can then be used to estimate spectral emissions. Rate or absorption rate, or temperature dependence of total emissivity or absorptivity. This information can then be used to improve the uniformity and reproducibility of the thermal process, and it is also possible to optimize the control of the heating process for maximum time efficiency and, therefore, optimal wafer throughput. Although the information about the type of wafer being processed can be separately supplied to the process equipment, sometimes it is not convenient or impossible to obtain such information. If this is the case, it may be necessary to determine the characteristics of the site (4), or immediately before the process or at an earlier stage of the process. For example, by taking an estimate of the internal transmittance (which is related to the absorption coefficient α(λ, τ)), we can understand the type of wafer being processed because the material properties of the substrate affect its absorption spectrum, which has been The wavelength dependence of α(λ, Τ) is described. In principle, the type of wafer being processed can be identified by measuring its absorption spectrum. For example, by comparing the measured absorption spectrum with the absorption spectrum data of a series of different materials, the material from which the substrate is made can be identified. For example, if the processed wafer is lightly doped and has a resistivity greater than that of 丨^^, its absorption spectrum shows a significant increase in absorption from the wavelength 〇·8μιη, at this time at room temperature. ') is about 85 = cm·1, and α(λ,τ) at room temperature is about 〇〇2 coffee·! until 1·2 μηη. This "bumped feature" is often referred to as the absorption limit, and its spectral position is related to the minimum energy gap strength in the band structure of the material. Identifying this in the absorption spectrum towel - a unique feature can help the touch wafer substrate is made of Shi Xi. Conversely, if the wafer is made of tantalum, the absorption limit characteristic will appear near the wavelength range of 18 calls. A similar approach can be used to distinguish other materials such as QaAs, ΐηρ, %/, GaSb, InN, SiC, and diamonds because the absorption spectrum of these materials also shows an absorption limit. This method can also be used to identify the presence of alloyed semiconductors, such as yttrium-niobium alloys, or quaternary alloys of GaAs and inp. The analysis of the absorption spectrum can even calculate the composition of these alloys, such as the ratio of Si to Ge. In principle, the analysis of α(λ, Τ) can also be used to distinguish between different types of isolation.

40 Eunice 20〇6\PU CASE\PU-〇68\PU-〇68-〇013\PU-068-0013-TSUEI.D〇C 1302358 The unique body and metal '(10) semiconductors are present's because of the big record (four) New features in the spectrum. For example, Xu recordings exhibit absorption characteristics as a result of electron transitions in the energy level. The characteristics are usually in the ultraviolet (uv), visible, and near-infrared rays of the electromagnetic spectrum; Many materials also exhibit absorption characteristics associated with the vibration of atomic species near their average position. This _ sex usually occurs at the infrared length test. Comparing the measured absorption spectrum with the reference information about the absorption limit, the electronic transition, and the vibrational absorption (4) spectral position, the ^ characteristic can be used to identify the material processed by the Lijiang River. The absorption of the county can affect the anti-light, degree And the transmitted light intensity, and both of them can be used to calculate the absorption spectrum I. In addition, if the surface is not transparent, the inverse __ analysis provides a similar signal spectrum that can also be used for the crystallization state. For example, the phenomenon of electron-active doping „,···· & is absorbed by free carriers. Free-carrier absorption is the result of the interaction of charged carriers with electromagnetic light white opaque semiconductor lattice. The strength is affected by the concentration of the sub-concentration. In the casting of wealth, from the domain of Heqi electronic money hole, according to the doping. In the -n type semiconductor, the main charge carrier is electron, and in the "P" (four) to make the electric complement is a hole. Free-carrier absorption (IV) Wavelength and temperature-related other: 2 = type estimation ° Such models can also include the addition of doping to the absorption spectrum, ^^^'image = energy gap in heavily substituted semiconductors, and energy bands The electronic jump. The information about the absorption spectrum is collected and analyzed by the -model, which can be "ϊΐΐϊϊ (f) or (d)) and its concentration. This message can then be used to predict the behavior of the oblique conductor lying in the manufacturing step. Available on the surface of the wafer H (4) The core can be coated on both the front and back sides. These coatings can be composed of: they can also have side cases to form a variety of component features, and the front, a 1 shape and other non-planar structures appear. Usually the component characteristics will be referred to as w before the wafer is described by =. The optical measurements described in this specification can also contribute to the properties of the near-planet surface. The resulting message can be used to improve the process control in the semiconductor 41 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 I. Discriminating the front and back reflections can help to more fully understand the film and construction on either surface of the wafer. Both the reflected and transmitted light measurements can be affected by the optical properties of both surfaces of the wafer, which will be further behind Discuss again. Compare the various reflections and The measured value of the illuminating light can help to identify whether the films on any surface are transparent or opaque at any known wavelength, and whether the substrate is transparent or opaque. For example, if the films on the wafer are opaque, That is to say, Tt=〇, if the wafer is coated with a metal layer, this may be the case, then the front reflectivity r*wf=r1wf and the higher order reflections R2WF, R3WF, etc. are all zero. In this case, the transmittance s * and all transmitted light components ΤΙ, T2, etc., are also zero. However, if the substrate is not opaque, that is, & gentry, and the back layer is not opaque, τ#〇, then the back reflectance is off RIwb and RlWB#〇. Therefore, comparing the reflected light components of the front and back illumination can help to identify whether the films on the surface of the wafer are opaque. If the surface is found to have the same reflectance and surface reflectance. However, this condition is not the same for the other surface, so it is very likely that the front surface includes a transparent film. It should be understood that if the back reflection ratio Rbb=〇, the front reflectance can be the same as its reflectance because There is a second reflection, but in this case the transmittance is not zero, unless the front happens to be a perfect reflector. This is unlikely to happen in practice. This method can also be used to find the front of the wafer. Or the opaque film behind. Other measured values are also analyzed in the same way. For example, if the transmittance s (or the first reflected light component T1) is zero and the back reflectance r*wb_r1wb, then we can infer that the front is opaque The right wafer substrate is opaque, then a=0 and the reflectivity of each surface of the wafer is equal to its reflectance 'that is, R*WF=Rtv and R*wB=Rbv. In addition, the reflected light component is in the stomach. And so on, and R2WB, R2WB, etc. are all zero. Further, the transmittance & and all the transmissions are intimately divided into two materials. Thus the analysis of this or the transmitted energy component can be inferred whether the substrate material is opaque at any known wavelength. However, it is conceivable that if the film on both surfaces is opaque (that is, if it is calving), even if the substrate is _ (ie, a judgment), __ (that is, R WF-Rtv, R WB) occurs. =Rbv, s*=〇, Τι=〇, τ2=〇, etc.) 42 1302358 Other tests that can be used to determine whether the wafer surface contains an absorptive film, is the reflectance of the radiation emitted from the panel, and The light shot from the outside of the substrate is the same, that is, whether Riv is off Rts, then the top surface of the wafer contains an absorptive film. In addition, the top surface contains an absorptive film. Similar rules can also be applied to describe the wafer. ~ The bottom film is transparent and the wafer is transparent. Then a special case is formed. /, the light is incident on the wafer from either side of the same reflectivity. At this time, the inflammation = Rm 1- (15), that is 'Whether wafer* and all other coatings are non-absorbent _simple test, check if R"wF= R*WB. By choosing to implement the series of waves g, you can honestly face the same phase. For example, by敎Infrared, «face, such as 丨.55 division, ^ out! = Display:: 2 absorption system, like big i ^, then the wafer should be heavily heterogeneous, like 疋 has a domain smaller than amem. The silk standard is based on the wavelength 岐, and its W is determined by the addition of 赖丝·卿县==. Fortunately, many wavelengths of actual _ quantity, face-to-face reflection hybridization, or some other navigation of the wrong record. If the test can also be implemented by a broadband source that covers a range of wavelengths. To calculate the value of α(λ,τ), it is usually necessary to calculate the internal transmittance using the formula 。. The internal transmittance can be reflected or transmitted by the image depicted in Figure 19, or can be measured from the WB surface. The value of the similar component ^ ^, know the other characteristics of the wafer. For example, rearrange the formula 7, "a" can be C: \ Eunice 2006 \ PU CASE \ PU-068 \ PU-068-W1 3 \ PU- 068-0013-TSUEI.doc 43 (16) 13〇2358

Therefore, in order to obtain the value of a, it is necessary to know rV, & Although = these quantities can be obtained from their _ magnitudes, the method described in Caixian ^ Nuclear Book = can help to reproduce the 3 rides, because the first surface of the bottle is obtained by the reflection of the light component of the hemp Rtv recorded. The reflectance RVp can also be obtained by the method of ageing Wei. In many practical examples, the methods disclosed in this specification can provide a complete description of the optical hybrids of interest, including more accurate internal penetration, and absorption coefficients α(λ, Τ). The internal transmittance can also be calculated from the measured value of the transmitted light. To rearrange the equation 6, the following equation 2R^(17) where 'in order to obtain the value of a' must know s*, Tt, Rfe, Tb, and & Similarly, these quantities can be obtained from other job values or calculations, but they are known to the county. The value of the internal transmittance can also be calculated from the measured value of the specific component of the reflected or transmitted radiation. For example, as described above, the intensity iT1 of the light ray 1 is known as . Therefore, the internal transmittance can be estimated from the following equation / (18) - Jxu T^bi The internal transmittance can be calculated from the intensity measurement value Ir2 of the light 'line R2, and its intensity is a2Tt2RbsI. Therefore, the internal transmittance can be estimated by the following formula a

(19) 44 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.doc 1302358 Internal transmittance can also be derived from the ratio of successive reflected or transmitted light, as Each of the transmitted or reflected rays emitted by the substrate is progressively attenuated compared to the previous one: the occurrence of this subtraction is the result of twice crossing the substrate and being reflected by the upper and lower faces. - Path beam attenuation - coefficient AtsRbs. Therefore, if _"intensity 疋Ir3, then the ratio K of the sum can be converted from the following formula by K = lR3 / lR2 = a2Rts.

K (20)

• A similar method can be used to calculate the ratio of successive transmitted light, such as τι to T2, etc. Radiation conditions The condition that the dome or the bottom surface does not contain an absorptive film on the main day is particularly noteworthy. this

^ t ΙΓίJ Tb=1_Rbv} Rtv=Rts r-=r- ° M : The value of the method can be _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Calculate ~ and . Count +£!£ξ3.)Χ 1 -

(21) _ The reflectivity of the wafer for the light incident from the lunar surface can be obtained by the following formula . ▲ (22} Transmittance is (23)

l^a2KK 45 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068^13-TSUEI.Doc 1302358 Γ Value judgment a (24) = the secret of the book It is true that the internal transmittance is determined by the smuggling. Reflectance ^ can be used to numerically

Rbv can be obtained by the method disclosed in Japanese. For example, by taking a reflection of the light: the reflected light ray can be used to derive the inverse two. Then, ίΓ. Including the crystal _ surface reflection - the secondary reflected light, can be pushed 1 eight, bv. Measure Rwf, R "port Rbv, calculate the internal transmission by Equation 24, and calculate the internal transmittance of the surface reflectance Ϊ s from the measured value of the transmittance s*. If the latter, the inner 卩 transmission is used. The following equations are obtained by rearranging the equation 23 to calculate a)2(Bu 4)2+2^/: - (25). It is recognized that once the transmittance s is different, the reflectances Rtv and Rbv are also calculated, and the equation can be used. 25 Calculate the internal transmittance. Similarly, the internal transmittance can be calculated from the measured value of the transmitted light T1 and the following equation for rearranging the formula 18 a Ιτι

(1UHK (26) The internal transmittance can also be obtained by measuring the intensity of the reflected light, and the following formula is rearranged by Equation 19 C:\£unice 2006\PU CASE\PU-068\PU-068-0013\PU- 068-0013-TSUEI.Doc l3〇2358 a L· (27) The internal transmittance can be obtained by measuring the ratio of the intensity of the transmitted or transmitted light and re-arranging with the formula 20 a

K to tv Han tv (28) _ In summary, the examples disclosed in the present specification allow for a more accurate determination of the absorption coefficient of the substrate material. In the case where the thin crucible is absorptive, further measurements or models may be required in order to obtain a sufficiently accurate absorption coefficient measurement. However, the case where the surface film is non-absorbent is practically important because it occurs in important practical applications, such as early in the component fabrication process, the oxidation or deposition process performed, and some on the semiconductor wafer. Some annealing processes, such as ion bump damage annealing. In these remnants, the surface is relatively transparent, at least for infrared wavelengths. In addition, in the case where a film having a pattern exists on the wafer, even if the film itself is absorptive, the wire indicates that the absorptive film only partially covers the surface of the wafer, thereby allowing a large amount of human touch. Partially transmitted into the substrate. It is conceivable that as long as the internal transmittance & and the transmittance in front of or behind the wafer are very small, even if the surface film does not show some absorption, the analysis of the non-absorbent film condition is still Can have a combination of _ quasi-hybrid. The ability of Wei to absorb any surface layer and/or surface coverage. Therefore, the method can be used even in the case of an absorbent film, such as a metal, a Wei compound or a heavily doped rotating body region, as long as the ancient features attract relatively less absorption than the wearing substrate to the other. - the surface of the "type", the wafer does not have a thick layer of absorbent material on its back side, so it is usually reasonable to assume that the back side is non-absorptive. The twentieth figure shows that the sheet material is at wavelength 4, Temperature phase of some optical properties 47

C:\Eunice 2006\PU

Ca^PU-〇6S\PU.〇6&.〇〇13\PU-068-0013- TSUEI. Doc 1302358 In this example, the absorption coefficient α of the material of the body portion of the plate (commonly referred to as the substrate) is manufactured. (λ, Τ) changes with temperature. Therefore, the internal transmittance also changes with temperature. In the following discussion, the methods disclosed in this specification are described more closely to illustrate how they are used to predict optical properties as a function of temperature. The method used relies on the low temperature measurement of the surface reflectance, coupled with the optical absorption model of 矽. In the example shown, a lightly doped germanium wafer is discussed. In this case, the light doping means that its resistivity is greater than about 1 Ω αη. The wafer thickness 775 _ discussed is a typical thickness for a 300 mm wafer fabricated for semiconductor components. In this example, the reflectance in front of the wafer is ~=0·3, and the reflectance in front of the wafer is Rbv=〇6, and it is assumed that the reflectance of the two surfaces does not change with temperature, and the surface does not There is an absorbent film. The optical characteristic is calculated as a function of temperature with a wavelength of 2.3 μm. In this example, the calculation is performed on a light shot that is incident perpendicular to the wafer or perpendicular to the wafer. The calculated values are the reflectivity r*wf in front of the wafer, the reflectivity r*wb behind the wafer, the wafer transmittance s*, the front of the wafer, the emissivity ewF, and the emissivity sWB behind the wafer. These values are derived from the combination of the above formulas 6, 7, 9 and 14 . All of these values are a function of wavelength and temperature, and in this example, the temperature dependence is a function of the _ partial transmittance as a function of temperature. The temperature dependence of the internal transmittance is due to the absorption system #i:a(vr). This specification shows the model proposed for α(λ, Τ), *VandenabeeleandMaexinJ Apple pyhs 72, 5867 (1992) with a wavelength of 2·3 pm. In general, the model used to derive the phenotype can be, for example, any theoretical or practical mode associated with the wavelength of the absorption coefficient of the material of the ten pairs of materials, such as 'for lightly doped stones. Said, R〇gne 吐 Α Α 卩 卩 卩 卩 卩 矽 矽 矽 矽 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四 四α(λ, Τ). In R〇ozeb〇〇m, "Fast,...W King and the progress of the integration process f (5 K } f 35 1 < ^ ^ # t ^ Timans such optical absorption of the second s and severely Model and data of rate and refractive index. ", for an estimate of a wide range of wavelengths, the visible infrared of λ about 0.5 μιη, said the infrared of peach i0rm. The above models can also take into account the substrate doping conditions,疋袷 Again, or other information about the concentration of electrons and holes in the substrate.

48 C:\Eunice 2006\PU CASE\PU-068\PU-068^13\PU.068-0013.TSUEI.DOC 1302358 The combined model is also described in the literature, including the Drude model, which is about free carrier absorption. It can be used to estimate the electron and hole concentration effects of infrared absorption. Other information needed to predict optical and thermal properties is the thickness of the wafer. The accuracy required for the predicted value, or the appropriate thickness of the wafer size to be processed (input by the user), or can be measured manually or manually. In the example depicted in the illustration, the α (λ, Τ) of 2.3 μη is very low at low temperatures, for example, it may be less than 10 〇 6 cm-i under the chamber. In this case, a is approximately 丨. In contrast, U is α (λ, Τ) large, such as about 1 〇〇 cm-1 at 730 °C. In this case a is approximately

0.00054. As the temperature increases, the internal transmittance approaches zero and the wafer becomes opaque. The figure shows that when the temperature is lower than about 250t: the internal transmittance is still about 丨, and the substrate is actually j, but as the temperature increases, the internal transmittance decreases until the temperature is greater than 75〇艽, the actual wafer It is opaque. At 25 〇 and 75 (rc interval, the wafer can be said to be translucent. At low temperatures, the wafer is transparent and the transmittance is s* = 〇.34, but this value approaches zero as the temperature increases. The temperature is greater than 75G °C depends on less than 1〇_4. In the low angle, the f-shoot rate of the light hitting the front and the shot _ light anti-theft _, as expected in the above formula is. In this fc example, 'normal temperature When R*wf=R*wb==〇66. However, when the temperature rises, & and r*wb^ are small and the sound is no longer the same. The decrease in value is due to the increase in the substrate_absorption, which is reduced. , (the other surface of the illuminated surface) the contribution of the reflected light to the refractive index. When the substrate becomes transparent, the reflection is converted to the reflectance corresponding to the illuminated surface, and the white (four) rate is zero. At this time, the wafer is transparent. The principle that the object cannot be absorbed and the light is not emitted is consistent with this phenomenon. When the temperature rises and the wafer becomes translucent, the emissivity increases, and the Lang wafer is actually opaque. The value is equal to the corresponding surface emission ratio, so ε·=1«7 and SwB=1_Rbu. 聿Green The invention can provide high temperature _ Wei report value, absorption of scale value: P R R R: Some quantities can be obtained during heating, for example by immediately in the process chamber in the f-range <ewF. However, in some cases it may be difficult to report the error in the fM L. Relatively speaking The present invention allows the determination of the wafer at any position outside the process chamber, such as 49 C.\Eumce 2006\PU ca^\PU-068\PU-068-0013\PU-068-0013-TSUEI.doc 1302358. A characteristic. So you can combine the knowledge about the trend of the temperature change and know the wafer thickness to predict the emissivity and yield of the wafer during the process. In this case, the measurement at room temperature is Sufficient to ensure that the wafer temperature I is greater than 75 (the exact value of the TC. The information of the emissivity can be used to correct a high temperature of 5 tens to determine the wafer temperature. The method of modeling can also be used to estimate the temperature function. The 0_ spectral absorption rate of the temperature side. This information can be provided to the control algorithm to improve the control of the heating process, # by proposing for the wafer and the lamp and teaching energy; Improved control' or by proposing about the radiant heat of the wafer: good

^第十十_流糊, shows how the class implements the present invention. The step is to place the day (four) field to a position where optical measurement is possible. The next step is to measure the optical properties. The starting temperature may be approximately room temperature ^ f ^ Lang Ren, miscellaneous. System_ riding crystal _-ΞΤ method. The next step is an optional step to determine the wafer thickness. This measurement can be two or two, measuring $. It is generally preferred to use a probe that does not touch the surface of the wafer to effect damage/contamination, especially in the case where electronic components are to be fabricated. For example, the thickness can be measured by an infrared interferometer. The thickness is also = ν, · U in the domain (four) face and in the latter case, the optical wavelength used by the wafer for the probe ϊί ==(4):::r can be based on laser triangulation. Wafer thickness = measured outside the air I position, measured by the _ surface position pair. The value can also be measured by the weight of the wafer, and the core coating itself is actually, 痄: in another example, if the wafer is included in the test*. This includes getting their thickness. The measured enthalpy of the two soil layers minus the thickness of the coatings can be determined by ===:' including measurement_doping. This measurement also obtained this message from the data of the Gebu people. If doping is required, usually 50 C:\Eunice 2006\PU CA ^E\PU-068\PU-068-0013\PU-068-0013-TSUEI. Doc 1302358 will require optical or electronic measurements. As described above, the method described in this measurement can be used to help determine the nature of the doping of the substrate. The information about the doping may include a doping type, for example, whether the wafer substrate is an n-type or p-type material. The value can also include the resistivity of the substrate. The value can also include the material used to dope the substrate, as well as the concentration of the dopant within the substrate. The value can also include the concentration of electrons or holes in the substrate. Other methods of determining doping may include direct electronic measurements with contact or non-contact probes. Non-contact probes are often preferred to avoid surface damage or contamination. The non-contact detection method may include sensing an eddy current induced by an oscillating electric field or a magnetic field applied to the wafer. Other messages can also provide information about the characteristics and characteristics of the wafer. For example, the information provided may include the nature of the wafer substrate, such as whether the wafer is germanium, kun, gallium, germanium, and the like. It may also include information about the nature of the film on the wafer, such as the thickness, material, and properties of the film on either side of the wafer. It can also include information about the rounds that appear on the surface of the wafer. Other properties that may be provided may also include thermal properties such as thermal conductivity, thermal diffusion rate, or defined heat capacity. The measurement of thickness and doping is considered selective here, as it is not necessary to know these too accurate values for some optical* predictions. However, wafer thickness measurements can also help to improve the various purposes of process control. For example, the thermal mass of a wafer is affected by its thickness. Therefore, the crystal heating or cooling rate is affected by the crystal thickness. Measurements of wafer thickness can help improve wafer heating or cooling control. For example, wafer thickness control can be provided to a control algorithm for setting the heating power. If the heating is an open circuit, for example, there is no feedback control due to temperature sensing and monitoring of the wafer temperature, which can also help control the process. This information can be used regardless of the type of heating used, and can be used to ride a heated wafer with electromagnetic waves or heat transfer by gas convection. For example, it can be used to improve the system in which the wafer is heated by a heating plate or stage. By accurately understanding the wafer thickness, it is easier to predict the temperature evolution after the wafer is placed on the - heater board. In this case, control improvements can be achieved even without measuring any optical characteristics of the wafer. These changes are mainly due to the fact that the wafers are mainly heated by heat transfer. In this case, the optical characteristics of the τ are better than the heat transfer from the wafer; there is a shirt, but the thermal quality of the wafer still plays a significant role in the heating cycle. The lower ν疋 model is used to predict at least one of the second temperatures of interest. 51 C:\Eunice 2006\PU CASE\PU-068\PU.〇68-0〇13\PU.〇68.〇013- TSUEI.Doc l3〇2358 sex T2. In practice, this may include predicting optical properties over a range of temperatures, such as establishing temperature dependence of optical properties. The optical properties may be any of the optical properties discussed in this specification, such as luminosity, absorptance, reflectance, transmittance or surface reflectance or emission ratio, 44. These characteristics can be predicted for any wavelength or temperature of interest. The type of enthalpy used may be based on the formula presented in this specification, or other set of formulas or algorithms that can predict optical properties. The data entered for the model includes any of the initial measured values performed at the first temperature τι. This can also optionally include money on wafer thickness and wafer doping. In the case where a wafer doping message is available, the optical absorption I coefficient and/or refractive index of the silk-screen substrate varies with wavelength and/or temperature. The next step involves using information about the optical properties to estimate and control the parameters associated with the heating process. Examples of parameters include the rate at which the wafer is used to sense the temperature of the wafer at the pyrometer. In this case, an improved emissivity estimate can provide a more accurate , reading. The pyrometer - typically measures the wafer temperature based on the measured radiation intensity values emitted by the wafer. The wafer emissivity or reflectance can be provided to an algorithm that calculates the wafer temperature based on the sensed base radiation intensity values. Many pyrometer solutions have been proposed for prior art towels. It is known to help reduce the temperature measured by the pyrometer by the emissivity of the emissivity by at least part of the wafer forming a reflective cavity to enhance the emissivity. However, if an initial estimate of the emissivity is available, it is possible to improve the correctness. Reduction. Diligence variation = other methods that should be used, including a geosonic optical measurement method used to measure wafer emissivity during the process. An example of this approach is the wave pyrometer approach. In such methods, the initial estimate of the emissivity can be used to improve the accuracy of the measurement. At this point, there is an important point to see the exact effect of stray light on the measured values. This type of light can be reflected by the wafer and then side-by-side with a pyrometer, introducing errors in temperature measurements. With an estimate of the reflectivity of the wafer, the reflected stray light component can be more accurately estimated, so its effects can be taken into account when measuring the temperature at the round. In addition, in the case where the wafer is translucent, it is often necessary to know both the wafer transmittance and the reflectance in order to determine the emissivity of the wafer, for example by Equation 8 or 13. The method of the present invention can also be used to determine transmittance and emissivity if desired. The measurement of transmission 2 can also help to obtain the amount of stray radiation transmitted by the substrate, which is taken into account in the sensing of the sensation to determine the temperature of the wafer. 52 C:\Eunke 2006\PU CASE\PU-068\PU-068-0〇13\PU-06a.〇〇n-TSUEi.Doc 1302358

^In some cases, emissivity and/or reflectivity can also be used to determine wafer temperature. For example, if you know the temperature dependence of any of these values at a known wavelength, you can use any of the local measurements. To determine the wafer temperature. The advantage of this approach is that it is no longer necessary to measure the radiation emitted by the wafer. This method can also be free from stray light problems. This method can also be applied to phase, lower temperature, which may be because the amount of heat radiation emitted is too low. Even if the coating that may be present on either surface of the wafer is not predictable, the temperature dependence of the reflectance or transmittance can be estimated using the methods described herein. For example, the reflectance of the front and back sides of the wafer can be obtained as above. The temperature dependence of the absorption coefficient can be obtained from a model as described above. The age t is measured and the measured reflectance is combined to provide the desired transmittance or reflectance temperature phase. Therefore, in this model, the parameter type is the temperature dependence of the rate or reflectivity. The parameter can also be used to determine the setting of the ageing female when the control is performed. This characteristic affects the energy transferred to the crystal, or the wafer, and therefore the wafer, temperature, heating rate or cooling rate. These magnitudes also affect the entire ^ specific area. Under the condition of the μ surface, the uniformity of the wafer temperature can be affected by the two systems (4). The additive characteristics may be process variables such as the power or energy delivered to heat the lamp, the temperature and position of the heating radiating element, the application of ▲ f-flow or voltage on a conductor, Rp or microwave intensity, or airflow intensity. Examples of other variables include: gas composition in the process chamber and its μ force, direction of gas flow, and the like. The position and size of the beam, the wavelength of the beam electromagnetic energy, the angle of incidence, the location of the heat source for the wafer orientation, the spacing between the wafer and the heating plate, or the spacing between the wafer and a heating bath, and so on. (4) The parameter of the method 'can be any factor that affects the thermal response of the wafer. i or system, the controller of this model 'for example, the algorithm can estimate the optimal process design; the produced: the sound is burned to know the wafer temperature of the heating cycle' and (or) maintain the wafer =rt :. These estimates can be based on better information on the = and two characteristics of the heat transfer phenomena that occur during the process, and it is possible to change the Bayesian and parallel to obtain better estimates of process or system variables. Control 53 C:\Eunke 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Dc 1302358 Less t-shirts provide models of phenomena to improve control; == prediction of values, For the purpose of including the sensor based on the model, the information may be included in the sensor setting, and the second part is in the crystal money TM - material position, the thickness of the wafer. : Rate. For example, the parameter can be to describe how the wafer is emitted: algorithm, or ^ can even be mixed. This information can be provided to control how and - energy II, the composition of the formula. For the same reason, if the upper limit is 4, the position and _ have been determined. It can also be selected based on the pre-measured data of the wafer characteristics, which can be switched between open loop and closed loop modes, for example, whether the reading is available at temperature. In law. The set temperature is reached during the period, the controller can be selected as a closed loop control film, and the redundancy type can include _ fire, crystallization, alloy, sintering, oxidation, and - two wafers 54 C:\Eunice 2006 \PU CASEXPl 'υ·〇68^υ.〇68.ωΐ3Ψυ-〇68.〇〇ΐ3.τ, 1302358 The last step is to take out the wafer. The scoop chart does not show another flow chart of the present invention. In this example, the flow chart clearly shows the feasibility of measuring the optical properties of a long and multiple wafer temperatures. The resulting interest from, for example, , can be used to estimate the optical properties of the wafer at process temperatures. Another flow chart shown in the twenty-third figure is clear and light, like the tenth in the nineteenth figure, wherein the light is only crystallized. In this case, the optical property measured may be crystal. The reflection of the circle front CWF) is Rb. In the display-example, where the measurement-surface reflectance is used to predict: the tenderness of the temperature scale. Wei miscellaneous values are available - very fresh model estimates, =Τ1=,. The emissivity can be used to correct the reading of a pyrometer. This value can also be used to Ά, and also source light energy. This value can also be used to estimate the number of wafer surface losses. The reflectance and emissivity measurements can be for a single-wavelength or a range of wavelengths. a second flowchart shows another flow chart description - an example in which the measured value is used to determine the crystal-doping characteristics. Then some of the optical properties at the estimated residual temperature = consider the doping bottle. Then turn this optical chick to the mosquito to monitor or control the parameters of $. This parameter may be determined—whether the pyrometer count is (4) the threshold temperature is accurate. This parameter may also be the wafer emissivity or absorptivity. This parameter may also be a marker, 4 control secrets (4) - a temperature measurement or control algorithm. It can also decide which temperature sensor to use for the column. If the wafer is measured to have a high height (inner/resistance is less than 〇·1 ΩΓΠ), the system can be selected to suit the known range of high temperatures. Measuring the temperature of the wafer. Conversely, if the wafer is determined to be lightly doped (for example, the resistivity is greater than 〇1), then a sensor can be selected to measure the wafer = degree based on the infrared light transmitted through the substrate. The determination of doping characteristics can also be used to improve the accuracy of temperature measurements. For example, if the infrared transmission measurement is used to determine the wafer temperature, then the wafer doping feature can be used to correct the effect of wafer doping on infrared transmission, and to obtain an accurate estimate of the wafer temperature. value.

55 ________•咖说(10) C 1302358 Another flowchart, shown in Figure 26, illustrates an example in which the thickness of the wafer is used to provide a parameter to the measurement or control system. This parameter may be the thickness of the film, such as '- a model-based control (4) can use this thickness message to estimate the wafer's rate of addition. The thickness message can also be used to estimate the time it takes for the wafer to reach a certain temperature. This is for 1 use. To improve the reproducibility of the heating process. The thickness of the crystal can be input into the process system, or the thickness of the system can be set in the system. Any of the flow chart methods or methods disclosed in the "Children's Month" can be combined as needed. These and other improvements of the present invention are difficult to achieve by the lining artisan = two will not deviate from the hair _ spirit and age. In addition, it is conceivable that Ke will implement the whole, or all or part of _. Moreover, the person skilled in the art should be able to understand that the above description is for the sake of example only, and is not intended to limit the invention. [Simple description of the drawing] The remaining parts of the specification (including the _ reference drawing) will More particularly and comprehensively, the present invention includes a preferred embodiment that can be understood by those skilled in the art, in which: Figure 1 is a side view of an embodiment of a thermal process chamber in which the method and system of the present invention may be used; Is a top view of an embodiment of the system according to the present invention; a third 岐-side view 'shows that the light beam is emitted onto the substrate, such as a semiconductor fourth image 疋 ' 显 显 显 显 显 显 显 显 显 显 显It is on the substrate of the swivel wafer; the body crystal =: side view 'shows that in another embodiment, two beams are emitted to the semi-conductor. The sixth figure is a side view of the implement that can be used with the __-light path. The seventh figure can be matched with σ. The other side of the light path is used as a side view of the light path; the figure / figure can be matched. The side view of the light path used in the present invention is a side view;

56 C:\Eunlce 2006\PU CASE\PU-068\PUO68-0013\PUO68-0013-TSUEI.DOC 1302358 The ninth figure is - the graphic 'shows the light intensity at different locations, as detailed below; It is a side view of an embodiment of the optical path that can be used in conjunction with the present invention. The tenth-figure is a side view of an embodiment of a light path that can be used in conjunction with the present invention. A side view of an embodiment of the present invention - a light path; a thirteenth view is a side view of an embodiment of the invention that can be used in conjunction with the present invention; and a fourteenth view is another light path that can be used in conjunction with the present invention. Side view of an embodiment

A side view of the fifteenth diagram showing a specific embodiment of the two-way irradiation of the wafer; a side view of the sixteenth embodiment showing another embodiment of the two-way irradiation of the wafer; An embodiment of illuminating a wafer at different locations in accordance with the disclosed method; a side view of the eighteenth view showing a beam of light being emitted onto a semiconductor wafer having front and back side cladding films; A side view of the nine figure showing the travel of light incident on the front surface of the wafer and the intensity of light at different locations;

The graph of Fig. 20 shows the temperature dependence of the optical characteristics of a wafer; and the method for measuring the characteristics of the wafer disclosed in the present invention by the eleventh through the twenty-sixth, the flow chart of the different embodiments . The reference symbols used in the specification and the illustrations are the same or analogous features or elements of the invention. 57 咖心2崎^ is said to be .06 age 06_chest·(10) side 1302358

[Main component symbol description] 10 System system 12 Processing chamber Process chamber 14 Wafer Wafer 15 Substrate holder Substrate bracket 16 Cooling conduit Cooling tube 18 Gas inlet Air inlet 20 Gas outlet Vent 21 Rotation mechanism Rotating mechanism 22 Heating device Heating Device 24 Lamp Set 25 Power Controller Power Controller 28 Light pipe Light Pipe 30 Light detector Photo Detector 50 System controller System Controller 100 Catridge Wa 110 Wafer handling device Wafer Transfer Device Wafer Optical Processing Chamber Wafer Optical Process Cavity 58 C:\Eunke 2006\PU CASE\PU-068\PUO68-0013\PU-068^0013*TSUEI.Doc

Claims (1)

1302358 X. Patent Application Range: 1. A method for controlling a substrate heating process, comprising: #射光光到- the first surface of the substrate, the substrate comprising the first surface and further separated from the first surface a second surface of thickness; directing the light into a path of light that separates the light reflected by the first surface from the light reflected by the second surface; detecting the amount of light reflected by the first surface; The amount of light reflected by the first surface is measured, and at least one system component in the process of heating the substrate is controlled or adjusted. 2. The method of claim 1, wherein the system component comprises a temperature measuring system comprising a radiation measuring device for sensing radiation emitted by the substrate during heating for determining a temperature of the substrate, The amount of light reflected by the first surface is measured to determine an emissivity of the substrate for use in conjunction with the amount of radiation sensed by the radiation measuring device for determining the temperature of the substrate. 3. The method of claim 1, wherein the system component comprises a heating system, comprising a heating device for heating the substrate, wherein the light from the first surface is measured for determining absorption of the substrate The rate is adjusted to adjust the power controller and thereby selectively increase or decrease the energy used to heat the substrate. 4. The method of claim 1, further comprising the steps of: an instrument/the amount of light reflected from the second surface of the substrate; and wherein the system component comprises a temperature measuring system comprising a radiation measuring device Sensing the amount of radiation emitted by the substrate during heating for determining a temperature of the substrate, the amount of light reflected by the first surface, and the amount of light reflected by the second surface being measured, The emissivity, transmittance, refractive index, or a combination thereof of the substrate is measured to be combined with the amount of light shot sensed by the Koda field measuring device to determine the temperature of the substrate. 5. The method of claim 1, wherein the substrate is added to a hot process chamber by adding 59 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013- T5UEI.Doc 1302358 heat' and the light is incident on the surface of the substrate outside the thermal processing chamber, and after detecting the amount of light reflected by the first surface, the substrate is moved into the thermal processing chamber. 6. If Shen Qing patent scope 帛! The method of the item, wherein the ray is below just. When c is incident on the first surface of the substrate. 7_ The method of claim 1, wherein the substrate is heated by a plurality of light sources, a heating bracket, radio frequency, microwave energy, hot wall environment, convection heating, conduction heating, energy during the heating process. The beam heating is like a plasma beam, an electron beam or an ion beam, or a combination of the above. 8. The method of claim 2, wherein the radiation measuring device senses a certain wavelength of radiation emitted by the substrate, and wherein the amount of light reflected by the first surface of the substrate is operated by a light-emitting measuring device Manufactured at the same wavelength, the light is incident on the first surface of the substrate at less than 1 〇 (rc). The method of claim 8 is the method of measuring the first surface of the substrate. The amount of light is used to determine the reflectance or emissivity of the substrate, and the temperature at the time of measurement is such that the transmittance of the substrate to the wavelength at which the radiation sensing device operates is less than 〇.1. The method wherein, in the heating process, the substrate is heated by electromagnetic radiation of a certain wavelength range, wherein the detection wavelength used by the light reflected and detected by the first surface of the substrate substantially comprises The method of heating the wavelength range of the electromagnetic wave of the substrate. The method of claim 5, wherein the system component comprises a temperature measuring system that senses the amount of radiation emitted by the substrate during heating for determining a temperature of the plate, the amount of light reflected by the first surface being measured to determine an emissivity of the substrate for use in conjunction with the amount of radiation sensed by the radiation measuring device for determining the temperature of the substrate And wherein the radiation measuring device detects the emission of a certain wavelength emitted by the substrate, and wherein the amount of light reflected by the first surface of the substrate is detected by the same wavelength of the radiation measuring device, the light Is the first surface of the substrate when it is less than 1 〇 (cm) ^ 60 (10) core fine 'P coffee side · _ side _ side as 1302358 12. The method of claim 11 of the patent scope The amount of light reflected by the first surface of the substrate and the amount of light reflected by the second surface of the substrate are used to determine the reflectance of the two sides of the substrate. The reflectance is used to determine the transmission of the substrate. Rate or emissivity, the temperature at which the measurement is such that the transmittance of the substrate to the wavelength at which the radiation sensing device operates is less than 0.1. 13. The method of claim 10, further comprising the step of detecting The amount of light reflected by the second surface of the substrate, the measured ray of light reflected by the first surface of the substrate, and the amount of light reflected by the second surface of the substrate are measured to determine the reflectance of both sides of the substrate The reflectance of each surface is used to determine the absorption of the substrate The rate at which the temperature is measured is such that the transmittance of the substrate for the wavelength range of the electromagnetic waves for heating the substrate is less than 0.1. The method of claim 1, wherein the optical path comprises at least two optical elements. 15. The method of claim n, wherein the at least two optical components comprise a first lens and a second lens. 16. The method of claim 3, wherein the optical path comprises a first optical element and a second optical element, directing light to a specific location on the first surface of the substrate, reflected by the first surface The light then passes through the third optical element again, from which the second optical element 'wire' reflects off the third optical element and contacts a fourth optical element for focusing on a photodetector. The method of claim 15, wherein the light reflected by the first surface is at least partially separated from the light reflected by the second surface by adjusting a focal length of the first lens and a focal length of the second lens. 18. The method of claim 14, wherein the light reflected by the first surface has at least a mouth P knife separated from the light reflected by the third surface by using a light blocking element. 19. The method of claim 1, wherein the light emitted by the towel onto the first surface of the silk plate comprises a laser beam. 20. If you apply for the patent item 14 method, find at least two optical components including 61 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 Mirror, mirror or a combination thereof. The method of claim 21, wherein the illuminating onto the first surface of the substrate is produced by a broadband source. 22. A method of determining at least one optical feature of a substrate, comprising: on a first surface of the substrate, the substrate comprising the first surface and a second surface spaced apart from the surface of the surface by a thickness; a light path that reflects the light reflected by the first surface and the amount of light reflected by the first surface by the second surface; and measures the amount of light reflected by the first surface, and determines the substrate at least - Optical characteristics, including the reflectance, emission ratio, impurity ratio or transmittance of the first surface of the substrate, or the reflectance, emissivity, absorptivity or transmittance of the substrate, or any combination of the above items 23.3=: The method of item 22, wherein the optical characteristic of the county plate is a method according to a specific 23 item, wherein the specific wavelength range includes a wavelength at which a radiation sensing element for measuring the degree of the μ substrate operates. 25. The method of item 23, wherein the wavelength range of the light substantially encompasses a range of wavelengths of electromagnetic radiation used to heat the substrate. 26. The method of item 22, wherein the female plate is at least - the optical characteristic is determined by the thermal two set two =: heating 'and at least - the system component is added 27.: =; =: method 'where the substrate is at least, Micro is an additive method in which the substrate includes a semiconductor wafer, and at least one system component of the system is based on at least an optical feature 62 C:\Eumce 2006\PU CASE\PU-068\ PU-068-0013\PU-068-0013-TSUEI.Doc 1302358 Control. 29. The method of claim 1, wherein the system component comprises a temperature measuring system that senses a quantity of radiation emitted by the substrate during heating for determining a temperature of the substrate, the measured The amount of light reflected by the first surface is used to determine an emissivity of the substrate for use in conjunction with the amount of radiation sensed by the radiation measuring device to determine the temperature of the substrate. 30. The method of claim 26, wherein the system component comprises a heating system comprising a heating device having a power supply controller for heating the substrate, the amount of light measured by the first surface being Used to adjust the power controller and thus selectively increase or decrease the energy used to heat the substrate. 31. The method of claim 26, further comprising the steps of: detecting an amount of light reflected by the second surface of the substrate; and wherein the system component comprises a temperature measuring system comprising a light-emitting measuring device Sensing the amount of radiation emitted by the substrate during heating for determining a temperature of the substrate, measuring the amount of light reflected by the first surface, and measuring the amount of light reflected by the second surface, The emissivity, transmittance, refractive index, or a combination thereof of the substrate is measured to be combined with the amount of radiation sensed by the radiation measuring device to determine the temperature of the substrate. 32. The method of claim 22, wherein the optical path comprises at least two optical elements. 33. The method of claim 30, wherein the at least two optical components comprise a first lens and a second lens. The method of claim 30, wherein the optical path comprises a first optical element and a second optical element, the light is directed to a specific position on the first surface of the substrate, and the light reflected by the first surface The receiver again penetrates the second optical element, the light reflecting the third optical element and contacting a fourth optical element for focusing on a photodetector. 63 C:\Eunice 2006\PU CASE\PU.〇68\PU.〇68.〇〇mPU.〇68-0013-TSUE,.Doc 1302358 35. The method of claim 32, wherein the first The light reflected by the surface is at least partially separated from the light reflected by the second surface by adjusting the focal length of the first lens and the focal length of the second lens. 36. The method of claim 32, wherein the light reflected by the first surface is at least partially separated from the light reflected by the second surface by using a shading element. 37. The method of claim 32, wherein the at least two optical elements comprise a lens, a mirror, or a combination thereof. 38. The method of claim 34, wherein the at least two optical elements comprise a lens, a mirror, or a combination thereof. 64 C:\Eunice 2006\PU CASE\PU-068\PU-068-0013\PU-068-0013-TSUEI.DOC